Abstract: People beings express themselves by giving opinions, feedback, suggestions or ideas about any object. Opinions can be expressed in many ways such as it can be expressed on twitter, facebook, reviews, blogs etc. World and technology is growing the faster rate so for taking decisions such as buying a product, voting for a politician etc. people are using opinion present on blogs, social networking sites or shopping websites. Nowadays, if someone has to buy a online product then he/she can first view other buyer’s reviews on the product site and take the right decision accordingly. There are many opinions on the web about an product so taking decisions might be difficult. Therefore, opinion mining is used for classifying the reviews according to their polarity. Opinion mining (OM) is a process of the mining opinion from the review. Analyzing customers review is more important for any user in making right purchasing decision product and organization. Opinion mining is also known as sentiment analysis. In this paper we survey on Opinion mining with respect to their different levels, architecture, techniques applied, tools used, comparative study of techniques and challenges.

Keywords: Opinions, Polarity, Opinion Mining, Sentimental Analysis, Data Mining, Web Mining.