Feature Extraction From Product Review Using Ontology

Abstract

Opinion mining is accepting more attention because of the development of blogs, e-commerce, news, reports, forums and additional web sources where individuals tend to express their opinions. Different people have different opinions. People’s thought may vary according to the domain and opinion may contain both positive and negative words. For a product, user may like or dislike some of its features. Filtering this review and extract domain related features is the important task of this paper. In this paper, ontology is used to extract the features and adjectives are used as the sentiment word. Sentiment Analysis is used to obtain positive or negative feature of the review.