label_sentiments.py

import pandas as pd  # For loading/saving CSV

# Step 1: Load the data
# Replace 'IMDB Dataset.csv' and 'review' with your actual file and column names
df = pd.read_csv('IMDB Dataset.csv')  # Example file name
texts = df['review'][:20]  # Use first 20 rows for testing

# Step 2: Prepare labels list
labels = []

print("Sentiment Annotation Tool")
print("Instructions: Label each text as 'positive', 'neutral', or 'negative'.")

for i, text in enumerate(texts):
    print(f"\nText {i+1}: {text[:100]}...")  # Show first 100 characters
    label = input("Enter label (positive/neutral/negative): ").lower().strip()
    while label not in ['positive', 'neutral', 'negative']:
        print("Invalid label. Try again.")
        label = input("Enter label (positive/neutral/negative): ").lower().strip()
    labels.append(label)

# Step 3: Add labels to DataFrame and save
df_labeled = df.copy()
df_labeled['sentiment_label'] = [''] * len(df_labeled)
df_labeled.loc[:len(labels)-1, 'sentiment_label'] = labels
df_labeled.to_csv('labeled_data.csv', index=False)

print("\nLabeled data saved to 'labeled_data.csv'! Ready for sentiment analysis or machine learning.")