Opinion Mining from Social Networks

Online social networks have become a popular communication tool for the masses. People are constantly sharing their opinions on social media. Therefore, social networking websites are a rich source for opinion mining. By analyzing the various opinions expressed on such sites, we can determine how well a product is doing in the current market. In this paper, we propose a system which uses the popular microblogging website Twitter for mining user opinions about products or services. Our proposed system presents an approach to extract data from Twitter and perform linguistic analysis on it. Combining the techniques of artificial intelligence and natural language processing, we attempt to classify the opinion as positive, negative or neutral. Using word dependencies and Part Of Speech tagging, we analyze data to track the opinions expressed on social media on the given subject. This paper proposes a system that can give the user information in the form of graphs and charts, about the opinions of the other users on the product or service.