Abstract: The advent of internet and World Wide Web the field of Sentiment Analysis is growing rapidly. There are numerous websites available on internet which provides analysis to users to give reviews about specific product. However the reviews expressed are mostly disorganized. An accurate method for sentiments could help us, to extract suggestions from the internet and predict customerís preferences which could prove valuable for Social Networks, Bulk suggestions and marketing research. There are various algorithms available for Sentimental analysis. Generally Sentimental analysis has three levels of granularities: Document level, Sentence level and Aspect level. In this paper, we study and analyze different issues, data sources, classification methods and evaluation metrics for Sentiment Analysis and text pattern fetching methods.

Keywords: Sentiment Analysis, Text classification, Machine Learning, Dataset, Lexicon Approach.