Sentiment Classification using Rough Set based Hybrid Feature Selection

Sentiment analysis means to extract opinion of users from review documents. Sentiment classification using Machine Learning (ML) methods faces the problem of high dimensionality of feature vector. Therefore, a feature selection method is required to eliminate the irrelevant and noisy features from the feature vector for efficient working of ML algorithms. Rough Set Theory based feature selection method finds the optimal feature subset by eliminating the redundant features. In this paper, Rough Set Theory (RST) based feature selection method is applied for sentiment classification. A Hybrid feature selection method based on RST and Information Gain (IG) is proposed for sentiment classification. Proposed methods are evaluated on four standard datasets viz. Movie review, product (book, DVD and electronics) review dataset. Experimental results show that Hybrid feature selection method outperforms than other feature selection methods for sentiment classification.