Opinion Mining/Sentiment Analysis

Abstract

There has been tremendous growth in the amount of user generated content on the Internet. This is due to rise in social networks and the embrace of web 2.0 or technologies. The Internet is no longer a place for consumption of information only, but a melting point of users interacting on various interests.

This project concentrates on a specific type of content – opinionated content. Several review sites exist where users can comment on their experience about a movie, product or service. The review sites allow users to rank their experience of such products or service.

The literature will be reviewed in detail. Several techniques for automatically analyzing such opinionated data will be explored. The focus will be on semantic orientation of the text. Semantic orientation is a measure of how far the opinion contained in the text differs from the group of other words surrounding the text. And will be implemented and compared with other similar work.