Automatic Product Aspect Identification for Opinion Mining

The growth of web 2.0 application, consumer feedback about product is analyzed to improve the quality of the product. The consumer feedback or reviews are extracted from the social media and then determine the polarity (positive, negative or objective) is called sentiment analysis. It is also known as opinion mining or appraisal extraction or review mining. The sentiment lexicon plays an important role in sentiment analysis. General purpose sentiment lexicons are not suitable for aspect level or identifying particular domain topics. It is also difficult to identify the polarity of same opinion word is used for different aspects for example, in the laptop review, “large” is negative for battery aspect while being positive for the screen aspect. The need for automatic tool to extract, analyze and also understand the consumer suggestions about a product in the aspect level. The novel method is used the context dependent sentiment lexicon. The lexicon is context dependent, domain dependent and also aspect dependent. Initially, the product aspects are extracted from the consumer feedback and then determine the opinion word related to the product aspect. Both aspect and the opinion word are given to the context dependent sentiment lexicon is used to determine the polarity of the opinion word. The novel method is more accurate than other state of the art methods.