Sentiment Analysis of movie reviews using SentiWordNet Approach

In this paper, a new kind of domain specific feature-based heuristic for sentiment analysis of movie reviews using aspect-level is presented. The unsupervised learning technique for sentiment classification is used. The SentiWordNet based scheme using two different linguistic feature selections containing adjectives, adverbs and verbs and n-gram feature extraction is performed. In aspect oriented algorithm analysis of textual reviews of a movie is performed and then a sentiment label on each aspect is assigned. By using sentiment labels score is generated. Such scores for each aspect from different reviews are aggregated. Finally, based on all parameters a net sentiment profile of the movie is generated. The documentlevel sentiment for each movie reviews using SentiWordNet approach is also computed. The generated sentiment profile of a movie is compared with the result derived by using document-level classification. The results obtained present that aspect-level classification generates a more accurate and focused sentiment profile than the document-level sentiment analysis.