Twitter sentiment analysis using various classification algorithms

Twitter is a web application built to find out what is happening at any instance through its micro blogging feature, anywhere in the world. Twitter posts are generally short and generated constantly by public and very well-suited for opinion mining. These messages can be classified as containing either positive or a negative sentiment on the basis of certain aspects with respect to a term based query. The past studies of sentiment classification are not very conclusive about which features and supervised classification algorithms are good for designing accurate and efficient sentiment classification system. We propose to combine many feature extraction techniques like emoticons, exclamation and question mark symbol, word gazetteer, unigrams to design more accurate sentiment classification system. This paper presents empirical comparison of six supervised classification algorithms.