Abstract: The ever growing usage of Web articles have resulted in change of web from read only to read write. This has led to the evolution of Sentiment Analysis, which is nothing but analysing the emotions of the given text. Sentiment analysis is a branch of Natural Language Processing (NLP). Sentiment analysis is done on the content such as online news, reviews, blogs, and tweets for social events, political movements, company strategies, marketing campaigns, product preference etc. This paper, will give the overall view of sentiment analysis. It also discusses about Common Knowledge Base and Common Sense Knowledge Base. Hour Glass Model is also discussed in the paper which is effectively being used for sentiment analysis. Finally, various approaches for sentiment analysis have been discussed and compared based on their advantages and disadvantages. It has been found that concept based approach is best suited for sentiment analysis, as it is applicable to multi-word expression.

Keywords: Natural Language Processing; Sentiment analysis; Hour Glass Model; Opinion Mining; Knowledge Base; Concept based approach.