Data mining through sentiment analysis: Lexicon based sentiment analysis model using aspect catalogue

摘要

Keeping in mind the peculiarity about product reviews, we have proposed a method that can help in determining the sentiment associated with the product in any review. Our method is based on `Lexicon based approach for Sentiment Analysis' with an additional feature of Aspect Catalogue. In our approach we first look for the keywords present in aspect catalogue to identify those sentences in which features of any product are mentioned. Taking into account only these sentences we further work on determining what opinion the speaker holds about features or aspects of the product. Aspect catalogue is referred again to find degree of importance corresponding to the feature with respect to the product. This is followed by calculation of numerical value of sentiment involved by taking the average of all the weightage of the features present in the text.