Socialism deliberative democracy is the important form our people’s democracy, and group discussion as an important channel for deliberative democracy, it is the realistic way for selecting a decision scheme that under public consensus. The malpractice of traditional offline large-scale group discussion is increasingly prominent, such as it is difficult to grasp discuss progress in real time, it can’t make the speak information structured and visualized, and it can’t provide correct guidance for participants, these questions badly affect the quality of discussion. Fortunately, along with the rapid development of the network information techniques, online discuss technology, user interest modeling technology, and personalized information recommendation technology emerge as the times require, it provides a platform for improving the quality of group discussion, and then it makes achieving deliberative democracy and improving the quality of decision possible.This paper analysis the relevant literature of group discussion, user interest modeling and personalized information recommendation at home and abroad, describes the rule of group discussion, and the steps and methods of user interest modeling and personalized information recommendation. Meanwhile, it analysis the existing method of text processing, text feature extraction, interest model representation, text clustering and semantic similarity calculation. Then we choice the words record of a hearing, modeling user interest model, and provide user with personalized service. At last, we put forward the idea of establishing a personalized information recommendation system of interest-based model, which can provide reference for developing of the information recommendation system. |