Document Categorization with Entropy Based TF/IDF Classifier

The task of text categorization is assigning a given text document to one or more predefined categories. High availability of digital data requires methods for automatic processing of this data. Day-by day increase of this digital data gives rise to the need of fast and better text classifiers. This paper mainly focuses on classifying data in context of text categorization. This paper reports a study conducted on 20 news group dataset, using TFIDF in the context of document categorization. Feature selection is added to this result to improvise the categorization. The results achieved using this algorithm are very promising when compared to conventional methods with features chosen on the basis of bag-of-words text.