Personalized modeling and sentiment analyzing on Chinese micro-blog posts

摘要

Trying to understand the sentiment of the human has always been the goal of artificial intelligence. But mining opinions and analyzing the sentiment information in internet is an ongoing challenge. The obstacles to the analysis of Chinese micro-blog are not only their very brief content, but also the lack of a suitable model of blogger personality to support it. This paper focuses on personalized modeling and sentiment analysis on Chinese Micro-blog posts. The main contribution of this paper is to take more comprehensive factors into consideration in the process of building personalized blogger model, which is used to assist and rectify the results calculated by sentiment analysis based on dynamic weighting adjustment. The experimental results also show the feasibility of the approach, and existing problems and future works are also present in the end.