Feature selection based classification of sentiment analysis using Biogeography optimization algorithm

Sentiment classification of social media has recently become popular among scientists due to the emergence of product reviews, blogs and social networking sites. A large number of reviews are difficult to evaluate personally. Moreover due to variable nature of reviews it becomes difficult, to compile overall result of reviews, to know which product is better than other. Researchers have already implemented machine learning techniques to analyze sentiment present in the given document. But execution time for these techniques increases due to the increase in feature set of data. Also irrelevant features participate in determining the sentiment of the given document, thereby varying the accuracy of the algorithm. In order to get much better classification, we propose a Biogeography based optimization algorithm to select optimal features set from given data. Then by using Naïve Bayes and Support Vector Machine techniques, we perform sentiment classification of product reviews. The proposed technique can be applied to other classification problems where feature set is large.