diff --git a/Node Classification & Link Prediction with GDSL - Marvel Data.ipynb b/Node Classification & Link Prediction with GDSL - Marvel Data.ipynb
new file mode 100644
index 0000000..9cf4ebc
--- /dev/null
+++ b/Node Classification & Link Prediction with GDSL - Marvel Data.ipynb	
@@ -0,0 +1,4744 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# What is node classification?\n",
+    "\n",
+    "Node classification is a supervised machine learning (ML) approach whereby existing nodes with known classes can be used to train a model that will learn the classes for nodes where they are unknown. In order to achieve this, the data must be split into two parts — a training graph and a testing graph — prior to predicting the classes for the unknown nodes. The training process involves splitting the training graph into two parts — a training and validation set — that will be used to fine tune the model through subsequent steps.\n",
+    "\n",
+    "Node classification is based on logistic regression. As in any ML model, care must be taken in choosing both the training and test sets as well as how to ensure that the model is not overfitting the data. In the case of the GDS Node Classification algorithm, an L2 norm is used as a penalty.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### ML with GDS ... continued from https://github.com/AliciaFrame/ML_with_GDS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define Neo4j connections\n",
+    "import pandas as pd\n",
+    "from neo4j import GraphDatabase\n",
+    "host = 'neo4j://localhost:7687'\n",
+    "user = 'neo4j'\n",
+    "password = 'letmein'\n",
+    "driver = GraphDatabase.driver(host,auth=(user, password))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def run_query(query):\n",
+    "    with driver.session() as session:\n",
+    "        result = session.run(query)\n",
+    "        return pd.DataFrame([r.values() for r in result], columns=result.keys())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Marvel Universe dataset\n",
+    "\n",
+    "In this example we will be using a dataset from the comics and movies associated with the Marvel Universe. This dataset can be found here. It contains 40,616 characters and 65,870 relationships connecting them. Additionally, the characters have numerous properties that can be associated with each node. We will be using this dataset to try and predict, off of a series of characters for training purposes, which characters are X-Men and which are not.\n",
+    "\n",
+    "### 1.  Data preparation\n",
+    "\n",
+    "We will begin by loading in the data to the database from a series of CSV files available online. This can be done with the following query:\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import_queries = \"\"\"\n",
+    "\n",
+    "CALL apoc.schema.assert({Character:['name']},{Comic:['id'], Character:['id'], Event:['id'], Group:['id']});\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroes.csv\" as row\n",
+    "CREATE (c:Character)\n",
+    "SET c += row;\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/groups.csv\" as row\n",
+    "CREATE (c:Group)\n",
+    "SET c += row;\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/events.csv\" as row\n",
+    "CREATE (c:Event)\n",
+    "SET c += row;\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/comics.csv\" as row\n",
+    "CREATE (c:Comic)\n",
+    "SET c += apoc.map.clean(row,[],[\"null\"]);\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroToComics.csv\" as row\n",
+    "MATCH (c:Character{id:row.hero})\n",
+    "MATCH (co:Comic{id:row.comic})\n",
+    "MERGE (c)-[:APPEARED_IN]->(co);\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroToEvent.csv\" as row\n",
+    "MATCH (c:Character{id:row.hero})\n",
+    "MATCH (e:Event{id:row.event})\n",
+    "MERGE (c)-[:PART_OF_EVENT]->(e);\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroToGroup.csv\" as row\n",
+    "MATCH (c:Character{id:row.hero})\n",
+    "MATCH (g:Group{id:row.group})\n",
+    "MERGE (c)-[:PART_OF_GROUP]->(g);\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroToHero.csv\" as row\n",
+    "MATCH (s:Character{id:row.source})\n",
+    "MATCH (t:Character{id:row.target})\n",
+    "CALL apoc.create.relationship(s,row.type, {}, t) YIELD rel\n",
+    "RETURN distinct 'done';\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroStats.csv\" as row\n",
+    "MATCH (s:Character{id:row.hero})\n",
+    "CREATE (s)-[:HAS_STATS]->(stats:Stats)\n",
+    "SET stats += apoc.map.clean(row,['hero'],[]);\n",
+    "\n",
+    "LOAD CSV WITH HEADERS FROM \"https://raw.githubusercontent.com/tomasonjo/blog-datasets/main/Marvel/heroFlight.csv\" as row\n",
+    "MATCH (s:Character{id:row.hero})\n",
+    "SET s.flight = row.flight;\n",
+    "\n",
+    "MATCH (s:Stats)\n",
+    "WITH keys(s) as keys LIMIT 1\n",
+    "MATCH (s:Stats)\n",
+    "UNWIND keys as key\n",
+    "CALL apoc.create.setProperty(s, key, toInteger(s[key]))\n",
+    "YIELD node\n",
+    "RETURN distinct 'done';\n",
+    "\"\"\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Graph import"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "with driver.session() as session:\n",
+    "    for statement in import_queries.split(';'):\n",
+    "        try:\n",
+    "            session.run(statement.strip())\n",
+    "        except:\n",
+    "            pass"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This query creates a series of nodes and their labels and properties: Comic, Character, Stats along with a variety of edges such as which comics the characters appeared in which comics, who is an enemy of whom, etc. You can see a schema of this graph using CALL db.schema.visualization() here:\n",
+    "\n",
+    "<img src=\"https://neo4j.com/developer/graph-data-science/_images/marvel-schema.png\" width=600 height=600 />"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. Move character traits to character nodes\n",
+    "We next bring in the character traits from the stats to be node properties. These node properties will eventually be used to build the node classification model via both embeddings as well as the tabular approach."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>count(c)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>470</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   count(c)\n",
+       "0       470"
+      ]
+     },
+     "execution_count": 61,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character)-[:HAS_STATS]->(s)\n",
+    "WITH c, s.strength as strength, s.fighting_skills as fighting_skills, s.durability as durability, s.speed as speed, s.intelligence as intelligence, s.energy as energy\n",
+    "SET c.strength=strength,\n",
+    "    c.fighting_skills=fighting_skills,\n",
+    "    c.durability=durability,\n",
+    "    c.speed=speed,\n",
+    "    c.intelligence=intelligence,\n",
+    "    c.energy=energy\n",
+    "RETURN count(c)\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 3. Create an appeared together relationship\n",
+    "Next, we set up the co-occurance of characters such that we can identify which characters appear with which other characters and how often (which will be used to identify the edge weighting). This is done via:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 64,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "MATCH (c1:Character)-[:APPEARED_IN]->(c:Comic)<-[:APPEARED_IN]-(c2:Character) \n",
+    "WITH c1, c2, count(c) as weight\n",
+    "MERGE (c1)-[:APPEARED_WITH{times:weight}]->(c2)\n",
+    "MERGE (c2)-[:APPEARED_WITH{times:weight}]->(c1)\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 4. one hot encode group membership <-- I ended up not using this, but useful to know how to "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "MATCH (group:Group)\n",
+    "WITH group\n",
+    "   ORDER BY group.name\n",
+    "WITH collect(group) AS groups\n",
+    "MATCH (c:Character)\n",
+    "WITH c, gds.alpha.ml.oneHotEncoding(groups, [(c)-[:PART_OF_GROUP]->(group) | group]) as group_membership\n",
+    "SET c.group_membership=group_membership\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Feature Engineering\n",
+    "Once our data is loaded in, it is time to start the process of engineering the features that will be used to populate our mode. For example, we might consider a variety of **centrality** measures of the character to be a feature that would be useful to train with. To obtain this, we first create an in-memory graph as:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 1. load graph with features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>degreeDistribution</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>database</th>\n",
+       "      <th>memoryUsage</th>\n",
+       "      <th>sizeInBytes</th>\n",
+       "      <th>detailSizeInBytes</th>\n",
+       "      <th>nodeProjection</th>\n",
+       "      <th>relationshipProjection</th>\n",
+       "      <th>nodeQuery</th>\n",
+       "      <th>relationshipQuery</th>\n",
+       "      <th>nodeCount</th>\n",
+       "      <th>relationshipCount</th>\n",
+       "      <th>density</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>modificationTime</th>\n",
+       "      <th>schema</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'p99': 765, 'min': 0, 'max': 1174, 'mean': 12...</td>\n",
+       "      <td>marvel-character-graph</td>\n",
+       "      <td>neo4j</td>\n",
+       "      <td>6288 KiB</td>\n",
+       "      <td>6439616</td>\n",
+       "      <td>{'relationships': {'total': 2704160, 'everythi...</td>\n",
+       "      <td>{'Person': {'properties': {'group_membership':...</td>\n",
+       "      <td>{'ENEMY_UNDIRECTED': {'orientation': 'UNDIRECT...</td>\n",
+       "      <td>None</td>\n",
+       "      <td>None</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>132845</td>\n",
+       "      <td>0.108896</td>\n",
+       "      <td>2021-03-24T13:12:40.608004000-04:00</td>\n",
+       "      <td>2021-03-24T13:12:41.079009000-04:00</td>\n",
+       "      <td>{'relationships': {'ENEMY_UNDIRECTED': {}, 'EN...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                  degreeDistribution               graphName  \\\n",
+       "0  {'p99': 765, 'min': 0, 'max': 1174, 'mean': 12...  marvel-character-graph   \n",
+       "\n",
+       "  database memoryUsage  sizeInBytes  \\\n",
+       "0    neo4j    6288 KiB      6439616   \n",
+       "\n",
+       "                                   detailSizeInBytes  \\\n",
+       "0  {'relationships': {'total': 2704160, 'everythi...   \n",
+       "\n",
+       "                                      nodeProjection  \\\n",
+       "0  {'Person': {'properties': {'group_membership':...   \n",
+       "\n",
+       "                              relationshipProjection nodeQuery  \\\n",
+       "0  {'ENEMY_UNDIRECTED': {'orientation': 'UNDIRECT...      None   \n",
+       "\n",
+       "  relationshipQuery  nodeCount  relationshipCount   density  \\\n",
+       "0              None       1105             132845  0.108896   \n",
+       "\n",
+       "                          creationTime                     modificationTime  \\\n",
+       "0  2021-03-24T13:12:40.608004000-04:00  2021-03-24T13:12:41.079009000-04:00   \n",
+       "\n",
+       "                                              schema  \n",
+       "0  {'relationships': {'ENEMY_UNDIRECTED': {}, 'EN...  "
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# list graphs already created\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.list()\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>graphName</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>marvel-character-graph</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                graphName\n",
+       "0  marvel-character-graph"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## drop all previous graphs in memory\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.list()\n",
+    "YIELD graphName AS namedGraph\n",
+    "WITH namedGraph\n",
+    "CALL gds.graph.drop(namedGraph)\n",
+    "YIELD graphName\n",
+    "RETURN graphName;\n",
+    "\"\"\")\n",
+    "\n",
+    "# run_query(\"\"\"\n",
+    "# CALL gds.graph.drop('marvel-character-graph')\n",
+    "# \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodeProjection</th>\n",
+       "      <th>relationshipProjection</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>nodeCount</th>\n",
+       "      <th>relationshipCount</th>\n",
+       "      <th>createMillis</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'Person': {'properties': {'group_membership':...</td>\n",
+       "      <td>{'ENEMY_UNDIRECTED': {'orientation': 'UNDIRECT...</td>\n",
+       "      <td>marvel-character-graph</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>132845</td>\n",
+       "      <td>90</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                      nodeProjection  \\\n",
+       "0  {'Person': {'properties': {'group_membership':...   \n",
+       "\n",
+       "                              relationshipProjection               graphName  \\\n",
+       "0  {'ENEMY_UNDIRECTED': {'orientation': 'UNDIRECT...  marvel-character-graph   \n",
+       "\n",
+       "   nodeCount  relationshipCount  createMillis  \n",
+       "0       1105             132845            90  "
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.create(\n",
+    "  'marvel-character-graph',\n",
+    "  {\n",
+    "    Person: {\n",
+    "      label: 'Character',\n",
+    "      properties: { \n",
+    "      strength:{property:'strength',defaultValue:0},\n",
+    "      fighting_skills:{property:'fighting_skills', defaultValue:0},\n",
+    "      durability:{property:'durability', defaultValue:0},\n",
+    "      speed:{property:'speed', defaultValue:0},\n",
+    "      intelligence:{property:'intelligence', defaultValue:0},\n",
+    "      group_membership:{property:'group_membership',defaultValue:[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]}\n",
+    "      }\n",
+    "    }\n",
+    "  }, {\n",
+    "    APPEARS_WITH_UNDIRECTED: {\n",
+    "      type: 'APPEARED_WITH',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      aggregation: 'SINGLE',\n",
+    "      properties: ['times']\n",
+    "    },\n",
+    "    APPEARS_WITH_DIRECTED: {\n",
+    "      type: 'APPEARED_WITH',\n",
+    "      orientation: 'NATURAL',\n",
+    "      properties: ['times'],\n",
+    "      aggregation: 'SINGLE'\n",
+    "    },\n",
+    "    ALLY_UNDIRECTED: {\n",
+    "      type: 'ALLY',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      aggregation: 'SINGLE'\n",
+    "    },\n",
+    "    ALLY_DIRECTED: {\n",
+    "      type: 'ALLY',\n",
+    "      orientation: 'NATURAL',\n",
+    "      aggregation: 'SINGLE'\n",
+    "    },    \n",
+    "    ENEMY_UNDIRECTED: {\n",
+    "      type: 'ENEMY',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      aggregation: 'SINGLE'\n",
+    "    },\n",
+    "    ENEMY_DIRECTED: {\n",
+    "      type: 'ENEMY',\n",
+    "      orientation: 'NATURAL',\n",
+    "      aggregation: 'SINGLE'\n",
+    "    }\n",
+    "   \n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "and then we use this graph to calculate the PageRank, Betweenness Centrality, and Hyperlink-Induced Topic Search (HITS) of each node and write those values back to the database:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. run centrality algos to add more features \n",
+    "### pageRank"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>67</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>20</td>\n",
+       "      <td>False</td>\n",
+       "      <td>{'p99': 4.413420677185059, 'min': 0.1499996185...</td>\n",
+       "      <td>33</td>\n",
+       "      <td>0</td>\n",
+       "      <td>171</td>\n",
+       "      <td>{'maxIterations': 20, 'writeConcurrency': 4, '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   writeMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0           67                   1105             20        False   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 4.413420677185059, 'min': 0.1499996185...                    33   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0            171   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'writeConcurrency': 4, '...  "
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "CALL gds.pageRank.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_DIRECTED'],\n",
+    "     writeProperty: 'appeared_with_pageRank'\n",
+    "});\n",
+    "\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>58</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>19</td>\n",
+       "      <td>True</td>\n",
+       "      <td>{'p99': 0.36827564239501953, 'min': 0.14999961...</td>\n",
+       "      <td>10</td>\n",
+       "      <td>0</td>\n",
+       "      <td>73</td>\n",
+       "      <td>{'maxIterations': 20, 'writeConcurrency': 4, '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   writeMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0           58                   1105             19         True   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 0.36827564239501953, 'min': 0.14999961...                    10   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0             73   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'writeConcurrency': 4, '...  "
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "CALL gds.pageRank.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_DIRECTED'],\n",
+    "     writeProperty: 'ally_pageRank'\n",
+    "});\n",
+    "\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>43</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>16</td>\n",
+       "      <td>True</td>\n",
+       "      <td>{'p99': 0.3315916061401367, 'min': 0.149999618...</td>\n",
+       "      <td>12</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1062</td>\n",
+       "      <td>{'maxIterations': 20, 'writeConcurrency': 4, '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   writeMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0           43                   1105             16         True   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 0.3315916061401367, 'min': 0.149999618...                    12   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0           1062   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'writeConcurrency': 4, '...  "
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "CALL gds.pageRank.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_DIRECTED'],\n",
+    "     writeProperty: 'enemy_pageRank'\n",
+    "});\n",
+    "\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### betweenness"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>60</td>\n",
+       "      <td>{'p99': 6057.406249880791, 'min': 0.0, 'max': ...</td>\n",
+       "      <td>161</td>\n",
+       "      <td>1</td>\n",
+       "      <td>409</td>\n",
+       "      <td>{'writeConcurrency': 4, 'writeProperty': 'appe...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  writeMillis  \\\n",
+       "0                   1105           60   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 6057.406249880791, 'min': 0.0, 'max': ...                   161   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             1            409   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 4, 'writeProperty': 'appe...  "
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// betweenness\n",
+    "CALL gds.betweenness.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_UNDIRECTED'],\n",
+    "     writeProperty: 'appeared_with_betweenness'\n",
+    "});\n",
+    "\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 93,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>69</td>\n",
+       "      <td>{'p99': 1849.0078048706055, 'min': 0.0, 'max':...</td>\n",
+       "      <td>28</td>\n",
+       "      <td>0</td>\n",
+       "      <td>10</td>\n",
+       "      <td>{'writeConcurrency': 4, 'writeProperty': 'ally...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  writeMillis  \\\n",
+       "0                   1105           69   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 1849.0078048706055, 'min': 0.0, 'max':...                    28   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0             10   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 4, 'writeProperty': 'ally...  "
+      ]
+     },
+     "execution_count": 93,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// betweenness\n",
+    "CALL gds.betweenness.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_UNDIRECTED'],\n",
+    "     writeProperty: 'ally_betweenness'\n",
+    "});\n",
+    "\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>59</td>\n",
+       "      <td>{'p99': 2607.9687480926514, 'min': 0.0, 'max':...</td>\n",
+       "      <td>42</td>\n",
+       "      <td>0</td>\n",
+       "      <td>11</td>\n",
+       "      <td>{'writeConcurrency': 4, 'writeProperty': 'enem...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  writeMillis  \\\n",
+       "0                   1105           59   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 2607.9687480926514, 'min': 0.0, 'max':...                    42   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0             11   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 4, 'writeProperty': 'enem...  "
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "// betweenness\n",
+    "CALL gds.betweenness.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_UNDIRECTED'],\n",
+    "     writeProperty: 'enemy_betweenness'\n",
+    "});\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### HITS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>90</td>\n",
+       "      <td>0</td>\n",
+       "      <td>9</td>\n",
+       "      <td>371</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  writeMillis  \\\n",
+       "0                   2210            201        False           90   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             9            371   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "CALL gds.alpha.hits.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>38</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>26</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  writeMillis  \\\n",
+       "0                   2210            201        False           38   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             1             26   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "CALL gds.alpha.hits.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>29</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>15</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  writeMillis  \\\n",
+       "0                   2210            201        False           29   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             1             15   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "CALL gds.alpha.hits.write('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 3. mutate the in-memory graph rather than reload\n",
+    "\n",
+    "We will also want these values added to the in-memory graph for the sake of calculating graph embeddings in the next step, which is achieved through the .mutate() command:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>20</td>\n",
+       "      <td>False</td>\n",
+       "      <td>{'p99': 4.413420677185059, 'min': 0.1499996185...</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0</td>\n",
+       "      <td>106</td>\n",
+       "      <td>{'maxIterations': 20, 'sourceNodes': [], 'rela...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   mutateMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0             0                   1105             20        False   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 4.413420677185059, 'min': 0.1499996185...                     6   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0            106   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'sourceNodes': [], 'rela...  "
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "\n",
+    "CALL gds.pageRank.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_DIRECTED'],\n",
+    "     mutateProperty: 'appeared_with_pageRank'\n",
+    "});\n",
+    "\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>19</td>\n",
+       "      <td>True</td>\n",
+       "      <td>{'p99': 0.36827564239501953, 'min': 0.14999961...</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0</td>\n",
+       "      <td>63</td>\n",
+       "      <td>{'maxIterations': 20, 'sourceNodes': [], 'rela...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   mutateMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0             0                   1105             19         True   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 0.36827564239501953, 'min': 0.14999961...                     2   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0             63   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'sourceNodes': [], 'rela...  "
+      ]
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "\n",
+    "CALL gds.pageRank.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_DIRECTED'],\n",
+    "     mutateProperty: 'ally_pageRank'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>16</td>\n",
+       "      <td>True</td>\n",
+       "      <td>{'p99': 0.3315916061401367, 'min': 0.149999618...</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0</td>\n",
+       "      <td>46</td>\n",
+       "      <td>{'maxIterations': 20, 'sourceNodes': [], 'rela...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   mutateMillis  nodePropertiesWritten  ranIterations  didConverge  \\\n",
+       "0             0                   1105             16         True   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 0.3315916061401367, 'min': 0.149999618...                     2   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0             46   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'maxIterations': 20, 'sourceNodes': [], 'rela...  "
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// pageRank\n",
+    "\n",
+    "CALL gds.pageRank.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_DIRECTED'],\n",
+    "     mutateProperty: 'enemy_pageRank'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>0</td>\n",
+       "      <td>{'p99': 6057.406249880791, 'min': 0.0, 'max': ...</td>\n",
+       "      <td>71</td>\n",
+       "      <td>0</td>\n",
+       "      <td>272</td>\n",
+       "      <td>{'nodeLabels': ['*'], 'sudo': False, 'relation...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  mutateMillis  \\\n",
+       "0                   1105             0   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 6057.406249880791, 'min': 0.0, 'max': ...                    71   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0            272   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'nodeLabels': ['*'], 'sudo': False, 'relation...  "
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// betweenness\n",
+    "\n",
+    "CALL gds.betweenness.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_UNDIRECTED'],\n",
+    "     mutateProperty: 'appeared_with_betweenness'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>0</td>\n",
+       "      <td>{'p99': 1849.0078048706055, 'min': 0.0, 'max':...</td>\n",
+       "      <td>35</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>{'nodeLabels': ['*'], 'sudo': False, 'relation...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  mutateMillis  \\\n",
+       "0                   1105             0   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 1849.0078048706055, 'min': 0.0, 'max':...                    35   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0              2   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'nodeLabels': ['*'], 'sudo': False, 'relation...  "
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// betweenness\n",
+    "\n",
+    "CALL gds.betweenness.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_UNDIRECTED'],\n",
+    "     mutateProperty: 'ally_betweenness'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>centralityDistribution</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1105</td>\n",
+       "      <td>0</td>\n",
+       "      <td>{'p99': 2607.9687480926514, 'min': 0.0, 'max':...</td>\n",
+       "      <td>39</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>{'nodeLabels': ['*'], 'sudo': False, 'relation...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  mutateMillis  \\\n",
+       "0                   1105             0   \n",
+       "\n",
+       "                              centralityDistribution  postProcessingMillis  \\\n",
+       "0  {'p99': 2607.9687480926514, 'min': 0.0, 'max':...                    39   \n",
+       "\n",
+       "   createMillis  computeMillis  \\\n",
+       "0             0              2   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'nodeLabels': ['*'], 'sudo': False, 'relation...  "
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "// betweenness\n",
+    "\n",
+    "CALL gds.betweenness.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_UNDIRECTED'],\n",
+    "     mutateProperty: 'enemy_betweenness'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>293</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  mutateMillis  \\\n",
+       "0                   2210            201        False             0   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             1            293   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "\n",
+    "CALL gds.alpha.hits.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['APPEARS_WITH_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>14</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  mutateMillis  \\\n",
+       "0                   2210            201        False             0   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             1             14   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 32,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "\n",
+    "CALL gds.alpha.hits.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ALLY_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>ranIterations</th>\n",
+       "      <th>didConverge</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2210</td>\n",
+       "      <td>201</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>14</td>\n",
+       "      <td>{'writeConcurrency': 0, 'writeProperty': '', '...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  ranIterations  didConverge  mutateMillis  \\\n",
+       "0                   2210            201        False             0   \n",
+       "\n",
+       "   postProcessingMillis  createMillis  computeMillis  \\\n",
+       "0                     0             1             14   \n",
+       "\n",
+       "                                       configuration  \n",
+       "0  {'writeConcurrency': 0, 'writeProperty': '', '...  "
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "\n",
+    "//HITS\n",
+    "\n",
+    "CALL gds.alpha.hits.mutate('marvel-character-graph',{\n",
+    "     relationshipTypes: ['ENEMY_DIRECTED'],\n",
+    "     hitsIterations: 50,\n",
+    "     authProperty: 'appeared_with_auth',\n",
+    "     hubProperty: 'appeared_with_hub'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 4. compute fastRP extended embedding (Fast Random Projection (FastRP) node embedding algorithm)\n",
+    "Lastly, we will use the Fast Random Projection (FastRP) embedding algorithm to create embedding vectors for each node, that will be used in one of our node classifications. Despite the fact that we will only be looking at a subset of this graph, namely X-Men and those who might be X-men or relate to them somehow, but we will compute the embeddings for the whole graph.\n",
+    "\n",
+    "#### I'm writing this back because I'm only going to train my model on known characters, but I want the embedding for the full graph"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.fastRPExtended.write('marvel-character-graph',{\n",
+    "    relationshipTypes:['APPEARS_WITH_UNDIRECTED'],\n",
+    "    featureProperties: ['strength','fighting_skills','durability','speed','intelligence','appeared_with_pageRank','ally_pageRank','enemy_pageRank','appeared_with_betweenness','ally_betweenness','enemy_betweenness','appeared_with_hub','appeared_with_auth'], //14 node features\n",
+    "    relationshipWeightProperty: 'times',\n",
+    "    propertyDimension: 45,\n",
+    "    embeddingDimension: 250,\n",
+    "    iterationWeights: [0, 0, 1.0, 1.0],\n",
+    "    normalizationStrength:0.05,\n",
+    "    writeProperty: 'fastRP_Extended_Embedding'\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 5. drop extra graphs\n",
+    "\n",
+    "Finally, we can drop the marvel-character-graph to free up some memory via CALL gds.graph.drop('marvel-character-graph')."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "run_query(\"\"\"\n",
+    "call gds.graph.drop('marvel-character-graph');\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Running the node classification algorithm"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Prior to the actual running of the node classification we must set up our training and testing graphs. There are a few things that we need to consider. First, we want to have roughly an equal number of X-Men to non-X-Men in our graph to prevent class imbalance. This means that first we will select all of the X-Men and set the property is_xman to identify these individuals:\n",
+    "### select-label-the-data-for-the-model.cypher"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 38,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 1. Select & label the data for the model and find the x-men and tag them, then flag to use in model\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character)-[:PART_OF_GROUP]-> (g:Group{name:'X-Men'})\n",
+    "SET c.is_xman=1, c:Model_Data;\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "\n",
+    "We see here that c.is_xman is set to an integer value of 1 above, which is required by the node classification algorithm to distinguish between the various classes.\n",
+    "\n",
+    "Next, we need to identify characters that are not X-Men. There are many more non-X-Men characters that appear with the X-Men, so we will **downsample** these through the requirement to have a degree greater than zero while also using a random number to determine whether that character will be put into the non-X-Men set and set their class to the integer value of 0:\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 39,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 2. find and include some unaffiliated individuals that are very far from x-men (but not orphan nodes) \n",
+    "## there are way more not x-men (133 with other affiliations, 936 with no known group) \n",
+    "## so we need to downsample for training\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character)\n",
+    "WHERE NOT (c)-[:PART_OF_GROUP]->(:Group) WITH c\n",
+    "WHERE NOT (c)-[:APPEARED_WITH*2..3]-(:Character{is_xman:1}) \n",
+    "AND apoc.node.degree(c)>0 WITH c\n",
+    "WHERE rand() < 0.2\n",
+    "SET c:Model_Data, c.is_xman=0;\n",
+    "\"\"\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, we will create a set of character that will be used for predictions after the model is trained:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 44,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 3. label the holdout data too (to predict on)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character)\n",
+    "WHERE NOT (c:Model_Data)\n",
+    "SET c:Holdout_Data;\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Once we have done this, we will create an in-memory graph encompassing these characters, their properties, and the class to be predicted.\n",
+    "### load-graph-for-class-prediction"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## drop graph for class prediction, if exists\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "call gds.graph.drop('marvel_model_data');\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## Drop Model ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.drop(\"xmen-model-fastRP\");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Observe that we have two character labels that are being put into the in-memory graph below, namely Character and Holdout_Character. This ensures that we are not mixing up the characters that will be used in the validation after the model is fully trained."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodeProjection</th>\n",
+       "      <th>relationshipProjection</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>nodeCount</th>\n",
+       "      <th>relationshipCount</th>\n",
+       "      <th>createMillis</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'Holdout_Character': {'properties': {'strengt...</td>\n",
+       "      <td>{'APPEARED_WITH': {'orientation': 'UNDIRECTED'...</td>\n",
+       "      <td>marvel_model_data</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>65612</td>\n",
+       "      <td>69</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                      nodeProjection  \\\n",
+       "0  {'Holdout_Character': {'properties': {'strengt...   \n",
+       "\n",
+       "                              relationshipProjection          graphName  \\\n",
+       "0  {'APPEARED_WITH': {'orientation': 'UNDIRECTED'...  marvel_model_data   \n",
+       "\n",
+       "   nodeCount  relationshipCount  createMillis  \n",
+       "0       1105              65612            69  "
+      ]
+     },
+     "execution_count": 45,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 2. load graph for class prediction\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.create(\n",
+    "  'marvel_model_data',\n",
+    "  {\n",
+    "    Character: {\n",
+    "      label: 'Model_Data',\n",
+    "      properties: { \n",
+    "        fastRP_embedding:{property:'fastRP_Extended_Embedding', defaultValue:0},\n",
+    "        //graphSAGE_embedding:{property:'graphSAGE_embedding', defaultValue:0},\n",
+    "        strength:{property:'strength', defaultValue:0},\n",
+    "        durability:{property:'durability', defaultValue:0},\n",
+    "        intelligence:{property:'intelligence', defaultValue:0},\n",
+    "        energy:{property:'energy', defaultValue:0},\n",
+    "        speed:{property:'speed', defaultValue:0},\n",
+    "        is_xman:{property:'is_xman', defaultValue:0}\n",
+    "      }\n",
+    "    },\n",
+    "    Holdout_Character: {\n",
+    "      label: 'Holdout_Data',\n",
+    "      properties: { \n",
+    "        fastRP_embedding:{property:'fastRP_Extended_Embedding', defaultValue:0},\n",
+    "        //graphSAGE_embedding:{property:'graphSAGE_embedding', defaultValue:0},\n",
+    "        strength:{property:'strength', defaultValue:0},\n",
+    "        durability:{property:'durability', defaultValue:0},\n",
+    "        intelligence:{property:'intelligence', defaultValue:0},\n",
+    "        energy:{property:'energy', defaultValue:0},\n",
+    "        speed:{property:'speed', defaultValue:0},\n",
+    "        is_xman:{property:'is_xman', defaultValue:0}\n",
+    "      }\n",
+    "    }\n",
+    "  }, {\n",
+    "    APPEARED_WITH: { //I don't actually need this for node classification\n",
+    "      type: 'APPEARED_WITH',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      properties: ['times'],\n",
+    "      aggregation: 'SINGLE'\n",
+    "    }\n",
+    "});\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "First let’s train a simple model that only uses some character properties for the training process.\n",
+    "### train-node-classifier-to-find-x-men-fast-rp\n",
+    "\n",
+    "\n",
+    "In the statement below, we are training a model based on the node properties of energy, speed, strength, durability, and intelligence. The targetProperty is the thing we are trying to solve for; in this case we are trying to determine the node property is_xman (1 for an X-Man, 0 for everyone else). The model will be able to return the weighted F1 score and the accuracy, but it is important to note that only the first metric is used for the actual training. We see that the validation set represents 20% of the test graph with 5-fold cross-validation. Finally, we set a series of parameters that are used to evaluate the model. In this case, we have provided a series of penalties using the L2 norm with a given number of training iterations. The training algorithm will identify the optimal model given these parameters, which is returned in the final portion of the query along with the training and test F1 weighted scores.\n",
+    "\n",
+    "When this is run on our dataset, we obtain the following results below:\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>winningModel</th>\n",
+       "      <th>trainGraphScore</th>\n",
+       "      <th>testGraphScore</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'maxIterations': 1000, 'penalty': 0.0625}</td>\n",
+       "      <td>0.44086</td>\n",
+       "      <td>0.284722</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                 winningModel  trainGraphScore  testGraphScore\n",
+       "0  {'maxIterations': 1000, 'penalty': 0.0625}          0.44086        0.284722"
+      ]
+     },
+     "execution_count": 48,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 4. compare to tabular properties\n",
+    "## if not using fastRP\n",
+    "\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.nodeClassification.train('marvel_model_data', {\n",
+    "   nodeLabels: ['Character'],\n",
+    "   modelName: 'xmen-model-properties',\n",
+    "   featureProperties: ['energy','speed','strength','durability','intelligence'], \n",
+    "   targetProperty: 'is_xman', \n",
+    "   metrics: ['F1_WEIGHTED','ACCURACY'], \n",
+    "   holdoutFraction: 0.2, \n",
+    "   validationFolds: 5, \n",
+    "   randomSeed: 2,\n",
+    "   params: [\n",
+    "       {penalty: 0.0625, maxIterations: 1000},\n",
+    "       {penalty: 0.125, maxIterations: 1000}, \n",
+    "       {penalty: 0.25, maxIterations: 1000}, \n",
+    "       {penalty: 0.5, maxIterations: 1000},\n",
+    "       {penalty: 1.0, maxIterations: 1000},\n",
+    "       {penalty: 2.0, maxIterations: 1000}, \n",
+    "       {penalty: 4.0, maxIterations: 1000}\n",
+    "       ]\n",
+    "    }) YIELD modelInfo\n",
+    "  RETURN\n",
+    "  modelInfo.bestParameters AS winningModel,\n",
+    "  modelInfo.metrics.F1_WEIGHTED.outerTrain AS trainGraphScore,\n",
+    "  modelInfo.metrics.F1_WEIGHTED.test AS testGraphScore\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The scores are low, but this is not surprising. We provided a very minimal number of properties on which to train the model, a problem that is compounded by the fact the the graph itself is quite small. So instead, let’s train a new model using the **FastRP embeddings**."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>modelInfo</th>\n",
+       "      <th>trainConfig</th>\n",
+       "      <th>graphSchema</th>\n",
+       "      <th>loaded</th>\n",
+       "      <th>stored</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>shared</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'modelName': 'xmen-model-properties', 'modelT...</td>\n",
+       "      <td>{'holdoutFraction': 0.2, 'params': [{'maxItera...</td>\n",
+       "      <td>{'relationships': {'APPEARED_WITH': {}}, 'node...</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "      <td>2021-03-24T15:27:03.669509000-04:00</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           modelInfo  \\\n",
+       "0  {'modelName': 'xmen-model-properties', 'modelT...   \n",
+       "\n",
+       "                                         trainConfig  \\\n",
+       "0  {'holdoutFraction': 0.2, 'params': [{'maxItera...   \n",
+       "\n",
+       "                                         graphSchema  loaded  stored  \\\n",
+       "0  {'relationships': {'APPEARED_WITH': {}}, 'node...    True   False   \n",
+       "\n",
+       "                          creationTime  shared  \n",
+       "0  2021-03-24T15:27:03.669509000-04:00   False  "
+      ]
+     },
+     "execution_count": 49,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## Drop Model ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.drop(\"xmen-model-properties\");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This is identical to our procedure above, however, we can see that we have replaced the featureProperties to be the **FastRP embeddings**. We would expect this model to perform better since the embedding process returns a **vector embedding** for each node that, in our case, is 250 elements long. In fact, we obtain the following results with the embeddings"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>winningModel</th>\n",
+       "      <th>trainGraphScore</th>\n",
+       "      <th>testGraphScore</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'maxIterations': 1000, 'penalty': 0.0625}</td>\n",
+       "      <td>0.929064</td>\n",
+       "      <td>1.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                 winningModel  trainGraphScore  testGraphScore\n",
+       "0  {'maxIterations': 1000, 'penalty': 0.0625}         0.929064             1.0"
+      ]
+     },
+     "execution_count": 50,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 3. train node classifier to find x-men: fastRP\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.nodeClassification.train('marvel_model_data', {\n",
+    "   nodeLabels: ['Character'],\n",
+    "   modelName: 'xmen-model-fastRP',\n",
+    "   featureProperties: ['fastRP_embedding'], \n",
+    "   targetProperty: 'is_xman', \n",
+    "   metrics: ['F1_WEIGHTED','ACCURACY'], \n",
+    "   holdoutFraction: 0.2, \n",
+    "   validationFolds: 5, \n",
+    "   randomSeed: 2,\n",
+    "   params: [\n",
+    "       {penalty: 0.0625, maxIterations: 1000},\n",
+    "       {penalty: 0.125, maxIterations: 1000}, \n",
+    "       {penalty: 0.25, maxIterations: 1000}, \n",
+    "       {penalty: 0.5, maxIterations: 1000},\n",
+    "       {penalty: 1.0, maxIterations: 1000},\n",
+    "       {penalty: 2.0, maxIterations: 1000}, \n",
+    "       {penalty: 4.0, maxIterations: 1000}\n",
+    "       ]\n",
+    "    }) YIELD modelInfo\n",
+    "  RETURN\n",
+    "  modelInfo.bestParameters AS winningModel,\n",
+    "  modelInfo.metrics.F1_WEIGHTED.outerTrain AS trainGraphScore,\n",
+    "  modelInfo.metrics.F1_WEIGHTED.test AS testGraphScore\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>modelInfo</th>\n",
+       "      <th>trainConfig</th>\n",
+       "      <th>graphSchema</th>\n",
+       "      <th>loaded</th>\n",
+       "      <th>stored</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>shared</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'modelName': 'xmen-model-fastRP', 'modelType'...</td>\n",
+       "      <td>{'holdoutFraction': 0.2, 'params': [{'maxItera...</td>\n",
+       "      <td>{'relationships': {'APPEARED_WITH': {}}, 'node...</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "      <td>2021-03-24T15:22:04.942774000</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           modelInfo  \\\n",
+       "0  {'modelName': 'xmen-model-fastRP', 'modelType'...   \n",
+       "\n",
+       "                                         trainConfig  \\\n",
+       "0  {'holdoutFraction': 0.2, 'params': [{'maxItera...   \n",
+       "\n",
+       "                                         graphSchema  loaded  stored  \\\n",
+       "0  {'relationships': {'APPEARED_WITH': {}}, 'node...    True   False   \n",
+       "\n",
+       "                    creationTime  shared  \n",
+       "0  2021-03-24T15:22:04.942774000   False  "
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## Drop Model, if you need to create other models ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.drop(\"xmen-model-fastRP\");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 5. Make some predictions!"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Using the FastRP model, let’s inspect some predicted nodes. To do this, we first have to run the prediction algorithm, which we will then write to the nodes themselves:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodePropertiesWritten</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2066</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>{'modelName': 'xmen-model-fastRP', 'predictedP...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   nodePropertiesWritten  mutateMillis  postProcessingMillis  createMillis  \\\n",
+       "0                   2066             0                     0             1   \n",
+       "\n",
+       "   computeMillis                                      configuration  \n",
+       "0              2  {'modelName': 'xmen-model-fastRP', 'predictedP...  "
+      ]
+     },
+     "execution_count": 51,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "##  1. lets predict node classes (aka: can we find more x-men?)\n",
+    "## Add the predictions to the in-memory graph\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.nodeClassification.predict.mutate('marvel_model_data', {\n",
+    "  nodeLabels: ['Holdout_Character'], //filter our the character nodes\n",
+    "  modelName: 'xmen-model-fastRP',\n",
+    "  mutateProperty: 'predicted_xman',\n",
+    "  predictedProbabilityProperty: 'predicted_xman_probability'\n",
+    "});\n",
+    "\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "... write to the nodes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>nodeProperties</th>\n",
+       "      <th>propertiesWritten</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>109</td>\n",
+       "      <td>marvel_model_data</td>\n",
+       "      <td>[predicted_xman, predicted_xman_probability]</td>\n",
+       "      <td>2066</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   writeMillis          graphName  \\\n",
+       "0          109  marvel_model_data   \n",
+       "\n",
+       "                                 nodeProperties  propertiesWritten  \n",
+       "0  [predicted_xman, predicted_xman_probability]               2066  "
+      ]
+     },
+     "execution_count": 52,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "##  1. lets predict node classes (aka: can we find more x-men?)\n",
+    "## Add the predictions to the in-memory graph\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.writeNodeProperties(\n",
+    "  'marvel_model_data',\n",
+    "  ['predicted_xman', 'predicted_xman_probability'],\n",
+    "  ['Holdout_Character']\n",
+    ");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We now look at some of the predictions for characters that are labeled as X-Men. To do this, we run the following query:\n",
+    "\n",
+    "**c.predicted_xman** returns the predicted class (in this case we are looking for characters that were labeled as X-Men by the model). The returned **c.predicted_xman-probability** gives the probability of each class, presented as a list where the first element is the probability of class 0 (not an X-Man) and the second element is the probability of class 1 (an X-Man). Our results will be as follows for the first returned character (with long alias list truncated for space):"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 157,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>c.name</th>\n",
+       "      <th>c.aliases</th>\n",
+       "      <th>c.predicted_xman</th>\n",
+       "      <th>c.predicted_xman_probability</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>Steven Rogers, Brett Hendrick, Buck Jones, Yeo...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.1387353529305271, 0.8612646470694726]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>James Buchanan Barnes</td>\n",
+       "      <td>James Buchanan Barnes, Captain America</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.1387253187111901, 0.8612746812888097]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Nick Fury (LEGO Marvel Super Heroes)</td>\n",
+       "      <td>null</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13878208115148838, 0.8612179188485113]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Sharon Carter</td>\n",
+       "      <td>Agent 13, Irma Kruhl, Fraulein Rogers, others</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13871332327426625, 0.8612866767257333]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Kate Bishop</td>\n",
+       "      <td>Hawkingbird, Mockingbird, Taskmistress, Weapon...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.1386900695302335, 0.8613099304697661]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>849</th>\n",
+       "      <td>Whizzer (Stanley Stewart)</td>\n",
+       "      <td>None</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13866771041539983, 0.8613322895845997]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>850</th>\n",
+       "      <td>Talon (Fraternity of Raptors)</td>\n",
+       "      <td>Lord Talon; impersonated Araki, and Smasher</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13864395710056085, 0.8613560428994388]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>851</th>\n",
+       "      <td>Lava-Man</td>\n",
+       "      <td>null</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13858830271278944, 0.8614116972872103]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>852</th>\n",
+       "      <td>Blue Blade</td>\n",
+       "      <td>null</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.1386942381917725, 0.8613057618082272]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>853</th>\n",
+       "      <td>Xavin</td>\n",
+       "      <td>null</td>\n",
+       "      <td>1</td>\n",
+       "      <td>[0.13886197684498922, 0.8611380231550104]</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>854 rows × 4 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                   c.name  \\\n",
+       "0                            Steve Rogers   \n",
+       "1                   James Buchanan Barnes   \n",
+       "2    Nick Fury (LEGO Marvel Super Heroes)   \n",
+       "3                           Sharon Carter   \n",
+       "4                             Kate Bishop   \n",
+       "..                                    ...   \n",
+       "849             Whizzer (Stanley Stewart)   \n",
+       "850         Talon (Fraternity of Raptors)   \n",
+       "851                              Lava-Man   \n",
+       "852                            Blue Blade   \n",
+       "853                                 Xavin   \n",
+       "\n",
+       "                                             c.aliases  c.predicted_xman  \\\n",
+       "0    Steven Rogers, Brett Hendrick, Buck Jones, Yeo...                 1   \n",
+       "1               James Buchanan Barnes, Captain America                 1   \n",
+       "2                                                 null                 1   \n",
+       "3        Agent 13, Irma Kruhl, Fraulein Rogers, others                 1   \n",
+       "4    Hawkingbird, Mockingbird, Taskmistress, Weapon...                 1   \n",
+       "..                                                 ...               ...   \n",
+       "849                                               None                 1   \n",
+       "850        Lord Talon; impersonated Araki, and Smasher                 1   \n",
+       "851                                               null                 1   \n",
+       "852                                               null                 1   \n",
+       "853                                               null                 1   \n",
+       "\n",
+       "                  c.predicted_xman_probability  \n",
+       "0     [0.1387353529305271, 0.8612646470694726]  \n",
+       "1     [0.1387253187111901, 0.8612746812888097]  \n",
+       "2    [0.13878208115148838, 0.8612179188485113]  \n",
+       "3    [0.13871332327426625, 0.8612866767257333]  \n",
+       "4     [0.1386900695302335, 0.8613099304697661]  \n",
+       "..                                         ...  \n",
+       "849  [0.13866771041539983, 0.8613322895845997]  \n",
+       "850  [0.13864395710056085, 0.8613560428994388]  \n",
+       "851  [0.13858830271278944, 0.8614116972872103]  \n",
+       "852   [0.1386942381917725, 0.8613057618082272]  \n",
+       "853  [0.13886197684498922, 0.8611380231550104]  \n",
+       "\n",
+       "[854 rows x 4 columns]"
+      ]
+     },
+     "execution_count": 157,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 3. check our predicted node classes\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character) \n",
+    "WHERE c.predicted_xman = 1 AND NOT c:Model_Data\n",
+    "RETURN c.name, c.aliases, c.predicted_xman, c.predicted_xman_probability \n",
+    "  \"\"\")\n",
+    "\n",
+    "## (some of the results are unlabeled x men, like Beast, others are agents of SHIELD (frequent antagonists) or allies (avengers))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Examining **Steve Rogers** further, he is not actually an X-Man. However, in the graph we can see that he has many :APPEARED_WITH relationships with actual X-Men, which can be seen via:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "If we were to explore other characters returned in this list, we would see that they also have several relationships with true X-Men. We also will note that there are actual X-Men who were not linked in the original data with the X-Men group that are really X-Men (for example: Beast, Cyclops, and Charles Xavier)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>c.name</th>\n",
+       "      <th>e</th>\n",
+       "      <th>x.name</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Cable (X-Men: Battle of the Atom)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Nuke</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Rogue (X-Men: Battle of the Atom)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Lockheed</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Sabretooth (House of M)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Sage</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Iceman (X-Men: Battle of the Atom)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Magik (Amanda Sefton)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Proudstar</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Beast (Earth-311)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Warren Worthington III</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Logan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Marrow</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Shadowcat (Age of Apocalypse)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Nightcrawler (Ultimate)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Professor X (Ultimate)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Banshee (Theresa Rourke)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Mystique (House of M)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Jubilee (Age of Apocalypse)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>19</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Forge (Ultimate)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Colossus (Ultimate)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Storm (Marvel Heroes)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>22</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Mimic</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Cyclops (X-Men: Battle of the Atom)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Jean Grey (Ultimate)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>Steve Rogers</td>\n",
+       "      <td>(times)</td>\n",
+       "      <td>Havok</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          c.name        e                               x.name\n",
+       "0   Steve Rogers  (times)    Cable (X-Men: Battle of the Atom)\n",
+       "1   Steve Rogers  (times)                                 Nuke\n",
+       "2   Steve Rogers  (times)    Rogue (X-Men: Battle of the Atom)\n",
+       "3   Steve Rogers  (times)                             Lockheed\n",
+       "4   Steve Rogers  (times)              Sabretooth (House of M)\n",
+       "5   Steve Rogers  (times)                                 Sage\n",
+       "6   Steve Rogers  (times)   Iceman (X-Men: Battle of the Atom)\n",
+       "7   Steve Rogers  (times)                Magik (Amanda Sefton)\n",
+       "8   Steve Rogers  (times)                            Proudstar\n",
+       "9   Steve Rogers  (times)                    Beast (Earth-311)\n",
+       "10  Steve Rogers  (times)               Warren Worthington III\n",
+       "11  Steve Rogers  (times)                                Logan\n",
+       "12  Steve Rogers  (times)                               Marrow\n",
+       "13  Steve Rogers  (times)        Shadowcat (Age of Apocalypse)\n",
+       "14  Steve Rogers  (times)              Nightcrawler (Ultimate)\n",
+       "15  Steve Rogers  (times)               Professor X (Ultimate)\n",
+       "16  Steve Rogers  (times)             Banshee (Theresa Rourke)\n",
+       "17  Steve Rogers  (times)                Mystique (House of M)\n",
+       "18  Steve Rogers  (times)          Jubilee (Age of Apocalypse)\n",
+       "19  Steve Rogers  (times)                     Forge (Ultimate)\n",
+       "20  Steve Rogers  (times)                  Colossus (Ultimate)\n",
+       "21  Steve Rogers  (times)                Storm (Marvel Heroes)\n",
+       "22  Steve Rogers  (times)                                Mimic\n",
+       "23  Steve Rogers  (times)  Cyclops (X-Men: Battle of the Atom)\n",
+       "24  Steve Rogers  (times)                 Jean Grey (Ultimate)\n",
+       "25  Steve Rogers  (times)                                Havok"
+      ]
+     },
+     "execution_count": 53,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "MATCH (c:Character {name: 'Steve Rogers'})-[e]->(x:Character)-[:PART_OF_GROUP]->(g:Group {name: 'X-Men'})\n",
+    "RETURN c.name, e, x.name\n",
+    "  \"\"\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## What is link prediction?\n",
+    "\n",
+    "Link Prediction is the problem of predicting the existence of a relationship between nodes in a graph. In this guide, we will predict co-authorships using the link prediction machine learning model that was introduced in version 1.5.0 of the Graph Data Science Library.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Clean up any existing graphs in memory"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>degreeDistribution</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>database</th>\n",
+       "      <th>memoryUsage</th>\n",
+       "      <th>sizeInBytes</th>\n",
+       "      <th>detailSizeInBytes</th>\n",
+       "      <th>nodeProjection</th>\n",
+       "      <th>relationshipProjection</th>\n",
+       "      <th>nodeQuery</th>\n",
+       "      <th>relationshipQuery</th>\n",
+       "      <th>nodeCount</th>\n",
+       "      <th>relationshipCount</th>\n",
+       "      <th>density</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>modificationTime</th>\n",
+       "      <th>schema</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'p99': 380, 'min': 0, 'max': 559, 'mean': 59....</td>\n",
+       "      <td>marvel_model_data</td>\n",
+       "      <td>neo4j</td>\n",
+       "      <td>3505 KiB</td>\n",
+       "      <td>3589376</td>\n",
+       "      <td>{'relationships': {'total': 540832, 'everythin...</td>\n",
+       "      <td>{'Holdout_Character': {'properties': {'strengt...</td>\n",
+       "      <td>{'APPEARED_WITH': {'orientation': 'UNDIRECTED'...</td>\n",
+       "      <td>None</td>\n",
+       "      <td>None</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>65612</td>\n",
+       "      <td>0.053784</td>\n",
+       "      <td>2021-03-24T15:20:31.595451000-04:00</td>\n",
+       "      <td>2021-03-24T16:26:37.099612000-04:00</td>\n",
+       "      <td>{'relationships': {'APPEARED_WITH': {'times': ...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                  degreeDistribution          graphName  \\\n",
+       "0  {'p99': 380, 'min': 0, 'max': 559, 'mean': 59....  marvel_model_data   \n",
+       "\n",
+       "  database memoryUsage  sizeInBytes  \\\n",
+       "0    neo4j    3505 KiB      3589376   \n",
+       "\n",
+       "                                   detailSizeInBytes  \\\n",
+       "0  {'relationships': {'total': 540832, 'everythin...   \n",
+       "\n",
+       "                                      nodeProjection  \\\n",
+       "0  {'Holdout_Character': {'properties': {'strengt...   \n",
+       "\n",
+       "                              relationshipProjection nodeQuery  \\\n",
+       "0  {'APPEARED_WITH': {'orientation': 'UNDIRECTED'...      None   \n",
+       "\n",
+       "  relationshipQuery  nodeCount  relationshipCount   density  \\\n",
+       "0              None       1105              65612  0.053784   \n",
+       "\n",
+       "                          creationTime                     modificationTime  \\\n",
+       "0  2021-03-24T15:20:31.595451000-04:00  2021-03-24T16:26:37.099612000-04:00   \n",
+       "\n",
+       "                                              schema  \n",
+       "0  {'relationships': {'APPEARED_WITH': {'times': ...  "
+      ]
+     },
+     "execution_count": 55,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "call gds.graph.list ();\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 69,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>graphName</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: [graphName]\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 69,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## drop graph for class prediction, if exists\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "call gds.graph.drop('marvel_model_data', false) YIELD graphName;\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>modelInfo</th>\n",
+       "      <th>trainConfig</th>\n",
+       "      <th>graphSchema</th>\n",
+       "      <th>loaded</th>\n",
+       "      <th>stored</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>shared</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'modelName': 'xmen-model-fastRP', 'modelType'...</td>\n",
+       "      <td>{'holdoutFraction': 0.2, 'params': [{'maxItera...</td>\n",
+       "      <td>{'relationships': {'APPEARED_WITH': {}}, 'node...</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "      <td>2021-03-24T16:24:23.219549000-04:00</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           modelInfo  \\\n",
+       "0  {'modelName': 'xmen-model-fastRP', 'modelType'...   \n",
+       "\n",
+       "                                         trainConfig  \\\n",
+       "0  {'holdoutFraction': 0.2, 'params': [{'maxItera...   \n",
+       "\n",
+       "                                         graphSchema  loaded  stored  \\\n",
+       "0  {'relationships': {'APPEARED_WITH': {}}, 'node...    True   False   \n",
+       "\n",
+       "                          creationTime  shared  \n",
+       "0  2021-03-24T16:24:23.219549000-04:00   False  "
+      ]
+     },
+     "execution_count": 65,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## List Models ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.list();\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 67,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>modelInfo</th>\n",
+       "      <th>trainConfig</th>\n",
+       "      <th>graphSchema</th>\n",
+       "      <th>loaded</th>\n",
+       "      <th>stored</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>shared</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'modelName': 'xmen-model-fastRP', 'modelType'...</td>\n",
+       "      <td>{'holdoutFraction': 0.2, 'params': [{'maxItera...</td>\n",
+       "      <td>{'relationships': {'APPEARED_WITH': {}}, 'node...</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "      <td>2021-03-24T16:24:23.219549000-04:00</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           modelInfo  \\\n",
+       "0  {'modelName': 'xmen-model-fastRP', 'modelType'...   \n",
+       "\n",
+       "                                         trainConfig  \\\n",
+       "0  {'holdoutFraction': 0.2, 'params': [{'maxItera...   \n",
+       "\n",
+       "                                         graphSchema  loaded  stored  \\\n",
+       "0  {'relationships': {'APPEARED_WITH': {}}, 'node...    True   False   \n",
+       "\n",
+       "                          creationTime  shared  \n",
+       "0  2021-03-24T16:24:23.219549000-04:00   False  "
+      ]
+     },
+     "execution_count": 67,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## Drop Model if it exist ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.drop(\"xmen-model-fastRP\");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 70,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>modelInfo</th>\n",
+       "      <th>trainConfig</th>\n",
+       "      <th>graphSchema</th>\n",
+       "      <th>loaded</th>\n",
+       "      <th>stored</th>\n",
+       "      <th>creationTime</th>\n",
+       "      <th>shared</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: [modelInfo, trainConfig, graphSchema, loaded, stored, creationTime, shared]\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 70,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## Drop Model ( for community limitations)\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.beta.model.list();\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Train and test datasets\n",
+    "### 1. Split the graph into the data we want to use for the model, and data to hold out to test afterwards"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 71,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 71,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 1. Split the graph into the data we want to use for the model, and data to hold out to test afterwards\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c1:Character)-[:APPEARED_IN]->(c:Comic)<-[:APPEARED_IN]-(c2:Character) \n",
+    "WHERE c.year <> \"2020\" AND c.year <> \"2019\" AND c.year <> \"2018\" AND c.year <> \"2017\" AND c.year <> \"2016\" \n",
+    "WITH c1, c2, count(c) as weight\n",
+    "MERGE (c1)-[:APPEARED_WITH_MODEL{times:weight}]->(c2)\n",
+    "MERGE (c2)-[:APPEARED_WITH_MODEL{times:weight}]->(c1);\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### And label the data that's been held out"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 72,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Empty DataFrame\n",
+       "Columns: []\n",
+       "Index: []"
+      ]
+     },
+     "execution_count": 72,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## And label the data that's been held out\n",
+    "run_query(\"\"\"\n",
+    "MATCH (c1:Character)-[:APPEARED_IN]->(c:Comic)<-[:APPEARED_IN]-(c2:Character) \n",
+    "WHERE c.year=\"2020\" OR c.year=\"2019\" OR c.year=\"2018\" OR c.year=\"2017\" OR c.year=\"2016\" \n",
+    "WITH c1, c2, count(c) as weight\n",
+    "MERGE (c1)-[:APPEARED_WITH_HOLDOUT{times:weight}]->(c2)\n",
+    "MERGE (c2)-[:APPEARED_WITH_HOLDOUT{times:weight}]->(c1);\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. load graph for class prediction"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>nodeProjection</th>\n",
+       "      <th>relationshipProjection</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>nodeCount</th>\n",
+       "      <th>relationshipCount</th>\n",
+       "      <th>createMillis</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'Character': {'properties': {'strength': {'pr...</td>\n",
+       "      <td>{'APPEARED_WITH_HOLDOUT': {'orientation': 'UND...</td>\n",
+       "      <td>marvel_linkpred_data</td>\n",
+       "      <td>1105</td>\n",
+       "      <td>54268</td>\n",
+       "      <td>50</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                      nodeProjection  \\\n",
+       "0  {'Character': {'properties': {'strength': {'pr...   \n",
+       "\n",
+       "                              relationshipProjection             graphName  \\\n",
+       "0  {'APPEARED_WITH_HOLDOUT': {'orientation': 'UND...  marvel_linkpred_data   \n",
+       "\n",
+       "   nodeCount  relationshipCount  createMillis  \n",
+       "0       1105              54268            50  "
+      ]
+     },
+     "execution_count": 73,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 2. load graph for class prediction\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.create(\n",
+    "  'marvel_linkpred_data',\n",
+    "  {\n",
+    "    Character: {\n",
+    "      label: 'Character',\n",
+    "      properties: { \n",
+    "        fastRP_embedding:{property:'fastRP_Extended_Embedding', defaultValue:0},\n",
+    "        //graphSAGE_embedding:{property:'graphSAGE_embedding', defaultValue:0},\n",
+    "        strength:{property:'strength', defaultValue:0},\n",
+    "        durability:{property:'durability', defaultValue:0},\n",
+    "        intelligence:{property:'intelligence', defaultValue:0},\n",
+    "        energy:{property:'energy', defaultValue:0},\n",
+    "        speed:{property:'speed', defaultValue:0},\n",
+    "        is_xman:{property:'is_xman', defaultValue:0}\n",
+    "      }\n",
+    "    }\n",
+    "  }, {\n",
+    "    APPEARED_WITH: { //I don't actually need this for node classification\n",
+    "      type: 'APPEARED_WITH_MODEL',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      properties: ['times'],\n",
+    "      aggregation: 'SINGLE'\n",
+    "    },\n",
+    "    APPEARED_WITH_HOLDOUT: { //I don't actually need this for node classification\n",
+    "      type: 'APPEARED_WITH_HOLDOUT',\n",
+    "      orientation: 'UNDIRECTED',\n",
+    "      properties: ['times'],\n",
+    "      aggregation: 'SINGLE'\n",
+    "    }\n",
+    "});\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 3. Add test train splits to in-memory graph"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>relationshipsWritten</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>51438</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   relationshipsWritten\n",
+       "0                 51438"
+      ]
+     },
+     "execution_count": 74,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 3. Add test train splits to in-memory graph\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.splitRelationships.mutate('marvel_linkpred_data', {\n",
+    "  relationshipTypes: ['APPEARED_WITH'],\n",
+    "  remainingRelationshipType: 'APPEARED_WITH_REMAINING',\n",
+    "  holdoutRelationshipType: 'APPEARED_WITH_TESTGRAPH',\n",
+    "  holdoutFraction: 0.2\n",
+    "}) YIELD relationshipsWritten;\n",
+    "  \"\"\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 75,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>relationshipsWritten</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>41153</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   relationshipsWritten\n",
+       "0                 41153"
+      ]
+     },
+     "execution_count": 75,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.splitRelationships.mutate('marvel_linkpred_data', {\n",
+    "  relationshipTypes: ['APPEARED_WITH_REMAINING'],\n",
+    "  remainingRelationshipType: 'APPEARED_WITH_IGNORED_FOR_TRAINING',\n",
+    "  holdoutRelationshipType: 'APPEARED_WITH_TRAINGRAPH',\n",
+    "  holdoutFraction: 0.2\n",
+    "}) YIELD relationshipsWritten;\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Model Training and Evaluation\n",
+    "### 4.  train a link prediction model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>winningModel</th>\n",
+       "      <th>trainGraphScore</th>\n",
+       "      <th>testGraphScore</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>{'maxIterations': 1000, 'penalty': 0.0}</td>\n",
+       "      <td>0.655248</td>\n",
+       "      <td>0.643368</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                              winningModel  trainGraphScore  testGraphScore\n",
+       "0  {'maxIterations': 1000, 'penalty': 0.0}         0.655248        0.643368"
+      ]
+     },
+     "execution_count": 76,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 4.  train a link prediction model\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.linkPrediction.train\n",
+    "('marvel_linkpred_data', {\n",
+    "  trainRelationshipType: 'APPEARED_WITH_TRAINGRAPH',\n",
+    "  testRelationshipType: 'APPEARED_WITH_TESTGRAPH',\n",
+    "  modelName: 'lp-appearance-model',\n",
+    "  featureProperties: ['fastRP_embedding'],\n",
+    "  validationFolds: 5,\n",
+    "  classRatio: 1.33,\n",
+    "  randomSeed: 2,\n",
+    "  params: [\n",
+    "    {penalty: 0.24, maxIterations: 1000},\n",
+    "    {penalty: 0.5, maxIterations: 1000},\n",
+    "    {penalty: 1.0, maxIterations: 1000},\n",
+    "    {penalty: 0.0, maxIterations: 1000}\n",
+    "  ]\n",
+    "}) YIELD modelInfo\n",
+    "RETURN\n",
+    "  modelInfo.bestParameters AS winningModel,\n",
+    "  modelInfo.metrics.AUCPR.outerTrain AS trainGraphScore,\n",
+    "  modelInfo.metrics.AUCPR.test AS testGraphScore\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 5. or  train a link prediction model - without an embedding"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## 5.  train a link prediction model - without an embedding\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.linkPrediction.train('marvel_linkpred_data', {\n",
+    "  trainRelationshipType: 'APPEARED_WITH_TRAINGRAPH',\n",
+    "  testRelationshipType: 'APPEARED_WITH_TESTGRAPH',\n",
+    "  modelName: 'lp-appearance-model-noEmbedding',\n",
+    "  featureProperties: ['strength','speed','intelligence','durability'],\n",
+    "  validationFolds: 5,\n",
+    "  classRatio: 1.33,\n",
+    "  randomSeed: 2,\n",
+    "  params: [\n",
+    "    {penalty: 0.24, maxIterations: 1000},\n",
+    "    {penalty: 0.5, maxIterations: 1000},\n",
+    "    {penalty: 1.0, maxIterations: 1000},\n",
+    "    {penalty: 0.0, maxIterations: 1000}\n",
+    "  ]\n",
+    "}) YIELD modelInfo\n",
+    "RETURN\n",
+    "  modelInfo.bestParameters AS winningModel,\n",
+    "  modelInfo.metrics.AUCPR.outerTrain AS trainGraphScore,\n",
+    "  modelInfo.metrics.AUCPR.test AS testGraphScore\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. lets predict some new links (aka: can we find more x-men?)\n",
+    "#### Add the predictions to the in-memory graph"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 77,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>relationshipsWritten</th>\n",
+       "      <th>mutateMillis</th>\n",
+       "      <th>postProcessingMillis</th>\n",
+       "      <th>createMillis</th>\n",
+       "      <th>computeMillis</th>\n",
+       "      <th>configuration</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>625</td>\n",
+       "      <td>{'modelName': 'lp-appearance-model', 'threshol...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   relationshipsWritten  mutateMillis  postProcessingMillis  createMillis  \\\n",
+       "0                  1000             0                     0             0   \n",
+       "\n",
+       "   computeMillis                                      configuration  \n",
+       "0            625  {'modelName': 'lp-appearance-model', 'threshol...  "
+      ]
+     },
+     "execution_count": 77,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## 2. lets predict some new links (aka: can we find more x-men?)\n",
+    "## Add the predictions to the in-memory graph\n",
+    "\n",
+    "run_query(\"\"\"\n",
+    "CALL gds.alpha.ml.linkPrediction.predict.mutate('marvel_linkpred_data', {\n",
+    "  relationshipTypes: ['APPEARED_WITH'], //filter out the known relationship type\n",
+    "  modelName: 'lp-appearance-model',\n",
+    "  mutateRelationshipType: 'APPEARED_WITH_PREDICTED',\n",
+    "  topN: 500,\n",
+    "  threshold: 0.49\n",
+    "});\n",
+    "\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>writeMillis</th>\n",
+       "      <th>graphName</th>\n",
+       "      <th>relationshipType</th>\n",
+       "      <th>relationshipProperty</th>\n",
+       "      <th>relationshipsWritten</th>\n",
+       "      <th>propertiesWritten</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>63</td>\n",
+       "      <td>marvel_linkpred_data</td>\n",
+       "      <td>APPEARED_WITH_PREDICTED</td>\n",
+       "      <td>None</td>\n",
+       "      <td>1000</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   writeMillis             graphName         relationshipType  \\\n",
+       "0           63  marvel_linkpred_data  APPEARED_WITH_PREDICTED   \n",
+       "\n",
+       "  relationshipProperty  relationshipsWritten  propertiesWritten  \n",
+       "0                 None                  1000                  0  "
+      ]
+     },
+     "execution_count": 78,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "run_query(\"\"\"\n",
+    "CALL gds.graph.writeRelationship(\n",
+    "  'marvel_linkpred_data',\n",
+    "  'APPEARED_WITH_PREDICTED'\n",
+    ");\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 4. check predicted links "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 79,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>c1.name</th>\n",
+       "      <th>c2.name</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>BLACKIE DRAGO VULTURE</td>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>SPECTRUM MONICA RAMBEAU</td>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>OOLA UDONTA</td>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>TANELEER TIVAN THE COLLECTOR Offbeat alien Tan...</td>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>UNREVEALED UNSPOKEN</td>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>995</th>\n",
+       "      <td>SIGYN</td>\n",
+       "      <td>Sprite</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>996</th>\n",
+       "      <td>SIGYN</td>\n",
+       "      <td>The Spike</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>997</th>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "      <td>The Spike</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>998</th>\n",
+       "      <td>MAXWELL \"MAX\" DILLON ELECTRO</td>\n",
+       "      <td>X-Ray (James Darnell)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>999</th>\n",
+       "      <td>SIGYN</td>\n",
+       "      <td>X-Ray (James Darnell)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>1000 rows × 2 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                               c1.name  \\\n",
+       "0                                BLACKIE DRAGO VULTURE   \n",
+       "1                              SPECTRUM MONICA RAMBEAU   \n",
+       "2                                          OOLA UDONTA   \n",
+       "3    TANELEER TIVAN THE COLLECTOR Offbeat alien Tan...   \n",
+       "4                                  UNREVEALED UNSPOKEN   \n",
+       "..                                                 ...   \n",
+       "995                                              SIGYN   \n",
+       "996                                              SIGYN   \n",
+       "997                       MAXWELL \"MAX\" DILLON ELECTRO   \n",
+       "998                       MAXWELL \"MAX\" DILLON ELECTRO   \n",
+       "999                                              SIGYN   \n",
+       "\n",
+       "                          c2.name  \n",
+       "0    MAXWELL \"MAX\" DILLON ELECTRO  \n",
+       "1    MAXWELL \"MAX\" DILLON ELECTRO  \n",
+       "2    MAXWELL \"MAX\" DILLON ELECTRO  \n",
+       "3    MAXWELL \"MAX\" DILLON ELECTRO  \n",
+       "4    MAXWELL \"MAX\" DILLON ELECTRO  \n",
+       "..                            ...  \n",
+       "995                        Sprite  \n",
+       "996                     The Spike  \n",
+       "997                     The Spike  \n",
+       "998         X-Ray (James Darnell)  \n",
+       "999         X-Ray (James Darnell)  \n",
+       "\n",
+       "[1000 rows x 2 columns]"
+      ]
+     },
+     "execution_count": 79,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "### 4. check predicted links \n",
+    "run_query(\"\"\"\n",
+    "MATCH (c1:Character)-[r:APPEARED_WITH_PREDICTED]->(c2:Character) \n",
+    "RETURN c1.name, c2.name\n",
+    "  \"\"\")"
+   ]
+  },
+  {
+   "attachments": {
+    "Screen%20Shot%202021-03-24%20at%208.49.08%20PM.png": {
+     "image/png": "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"
+    }
+   },
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "![Screen%20Shot%202021-03-24%20at%208.49.08%20PM.png](attachment:Screen%20Shot%202021-03-24%20at%208.49.08%20PM.png)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}