| Microblog is platform for information distribution, capture, sharing and transmission, it provides one-way attention mechanism, the user can choose to focus on other users, and the users can automatically receive these users tweets, but don’t require these user concerns him. These concerned users may be one of interest to the user, such as colleagues, friends, relatives, or celebrities. The convenience of this way information dissemination, focused attention of social autonomy and extensive connections, make this social new media microblog popular with Internet users, the new media become a most important ways to user get the information and learn the social dynamics.Microblog information is usually recorded user behavior, which inevitably reflects the user’s interests and hobbies. By microblog data mining and analysis, can be found the user’s interest. For each user’s specific interest, can recommend the relevant information, products, advertising, etc, and also can recommend their users that might be interested users, so as to achieve the effect of personalized recommendations. In addition, plays an important role to enhance customer satisfaction for website and and promote the website development of itself. Therefore, the study found that microblog users interested in personalized recommendation and improve user satisfaction with the microblogging social network has great significance and value.To get an accurate Weibo user interest model to study the idea of this paper is to analyze the microblogging and microblogging features users get an accurate description of microblogging users interested in information, and then gives the user applies Weibo user interest modeling method Finally, the modeling methods tested in this paper. In this paper, Sina microblogging, for example, the main research work includes:In order to get accurate microblog user interest model, our research ideas in this paper is to get an accurate description of microblog users interested information by analyze the characteristics of microblog and microblog users, then give a suitable users interest modeling method for microblog users,and finally tested of this modeling method in this paper. We take sina weibo as an example in this paper, the main research work includes:1.Analyzing the characteristics of Sina microblog platform and user information of Sina microblog, conducted investigation in order to understand the correlation between microblog user information and user interest,and obtained the information that may indicate microblog user’s interested. Through statistical analysis found microblog text content and user attention interest can more comprehensive and accurate performance of the user’s interest,so in this paper we select the users posted content information and the particular users which microblog users concerned(celebrities and content classes microblog users) as the information sources of microbolg user interest.2.Combining the characteristics of microblog users interests information, gives the basic framework and methodology of microblog users interest modeling and design experiments to verify the feasibility of the method we proposed, and proposed the evaluation index to validates the accuracy of the microblog users interest model that we get in this paper.The experimental results show that this method can through the verification of the evaluation index, and can found microblog user?s interest more effectively and accurately. Finally update the user interest model based on the memory curve. |