Information Retreival project done as a part of the course CSF469.
Aim of the project is to find a fast feature selection technique(as compared to chi squared), with a better or equivalent performance in terms of F1-score.
WebKb and classic(cacm,cisi,cran,med) datasets have been used for training and testing purposes.Linear SVM classifier has been used for classification of document classes.
Packages used(Python 2.7.12):
Anaconda2 4.1.1
stemming 1.0.1
nltk 3.2.1