Sentiment analysis, also known as opinion mining, is the analysis of the feelings (i. e. attitudes, emotions and opinions) behind the words. Sentiment analysis involves classifying the opinions as positive, negative, or neutral. Classification of textual objects in accordance with sentiment is considered to be a more difficult task than classification of textual objects in accordance with the content because opinions in natural language can be expressed in subtle and complex ways containing slang, ambiguity, sarcasm, irony, and idiom. This paper investigates the problem of sentiment analysis of online review. A Jaccard index based clustering algorithm (JIBCA) is proposed to support mining online reviews and predicting sales performance. The information gain is the change in information by considering number of datasets. The performance of information gain varies depending on the dataset. It is observed that the information gain performed better in JIBCA than existing methods for the movie review dataset. It is therefore recommended that JIBCA can be a good feature selection method for sentiment classification tasks. This paper also proposes a new approach for movie reviews classification based on extraction and analysis of appraisal groups such as action, thrill, comedy, and romantic.

Sentiment Analysis Review Classification Opinion Mining Review Mining Prediction.