Movie Reviews Clustering Research Based On Opinion Target Extracting

As the development of the society,the Internet has gradually entered people's life,playing an indispensable role in daily life.People make comments on a variety of social platforms for different purposes.As a representative of the social platform,microblogging brings a huge number of comments.How to extract sentimental information from the Microblogs has become one of the hot topics in the current research.In this paper,the main research target is reviews of movie and TV series,in order to establish a system to extract review target and cluster them automatically.In the first two chapter,In the first two chapters,the related works are introduced,including evaluation target extracting,polarity judgment and clustering,and then put forward the main research contents and solutions in chapter 3.System implement and feature design are described in chapter 4.In the end,this method is compared to several mainstream methods,evaluation results show that our method is effective In the 6th COAE,our system ranked 1 st in product name extraction task and 6th for the whole task 5.