JIBCA: Jaccard Index based Clustering Algorithm for Mining Online Review

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