Detection of web users' opinion from normal and short opinionated words

Abstract

In this paper we present an approach to identify opinion of web users from an opinionated text and to classify web user's opinion into positive or negative. Web users document their opinion in opinionated sites, shopping sites, personal pages etc., to express and share their opinion with other web users. The opinion expressed by web users may be on diverse topics such as politics, sports, products, movies etc. These opinions will be very useful to others such as, leaders of political parties, selection committees of various sports, business analysts and other stake holders of products, directors and producers of movies as well as to the other concerned web users. Today web users express their opinion using normal words and short words. These short words, such as gud for good, grt8 for great etc., are very popular and are used by a large number of web users to document their opinion. We use semantic based approach to find users opinion from both normal and short words. Our approach first detects subjective phrases and uses these phrases along with intensifiers and diminishers to obtain semantic orientation scores. The semantic orientation score of these phrases is used to identify user's opinion from an opinionated text. Our approach provides better results than the other approaches on different data sets. © 2012 Springer-Verlag.