Research On WEB Knowledge Push Based On Multi - Dimensional Social Network In Social Tagging System

In the Web2.0 era, the Internet has become the world's largest knowledge base to bring great convenience to the people, but also led to the "knowledge lost", "knowledge overflow" and other issues. In recent years the rise of Web knowledge push can solve these problems, based on the records of Web user behavior on the Internet, mining the knowledge needs of Web users, and then integrate related resources or knowledge offered to the user, to meet individual needs. In this paper, we try to explore the model in the field of multi-dimensional social network analysis in order to optimize the effect of Web knowledge push.This paper first elaborates the theory of socialized labeling system, social network analysis and Web knowledge push from the aspects of connotation, characteristic, model and function. Then, the paper puts forward the construction method of multi-dimensional social network, and scientifically "transplant" to explore the multi-dimensional social network analysis framework of individual level, local level and global level by referring to the single - dimensional social network analysis principle. Secondly,it explores the connotation of multi-dimensional social network based on label subject,and analyzes the analysis index by multi-dimensional social network analysis. Thirdly,we build the Web knowledge push model based on the multi - dimensional social network, calculate the interest degree of the Web users' knowledge resources by using the multi - dimensional community discovery result and the collaborative filtering idea, and finally generate the Web knowledge push set. Finally, combining with the real data sets,construction and analysis of the multi-dimensional social networks based on the theme of the label, and on the basis of Web knowledge push model achieves Web knowledge push,through the accuracy evaluation proving the feasibility and effectiveness of the method proposed in this paper.