Abstract:
In consideration of the impact of user's interests, trust transfer, time factor, social network on personalized recommendation results in micro-blog contents, a new method is proposed in this paper. The method is based on community discovery, and recommends personalized micro-blog contents for users by improving the user model method, effectively using the social network of micro-blog platform and optimizing the recommended micro-blog's utility function. By comparing with other algorithms on the real micro-blog data, this method has a good effect of recommendation, has a higher accuracy and novelty, and analyzes the influence of the selection of parameters on the recommendation effect.