MMR-based feature selection for text categorization

Abstract

We introduce a new method of feature selection for text categorization. Our MMR-based feature selection method strives to reduce redundancy between features while maintaining information gain in selecting appropriate features for text categorization. Empirical results show that MMR-based feature selection is more effective than Koller & Sahami’s method, which is one of greedy feature selection methods, and conventional information gain which is commonly used in feature selection for text categorization. Moreover, MMR-based feature selection sometimes produces some improvements of conventional machine learning algorithms over SVM which is known to give the best classification accuracy.