Research On User-based Analysis Method In Social Tagging System

The term social tagging is currently used to indicate a support data structure that allows human users to classify and categorize various kinds of resources (e.g., URLs, photos, videos, and so on) by means of tags. The tags are made of the users, representative of the users'views and the reflection of the users' preference.Based on it, social tagging has some strong points, such as least-effort retrieval, low cognitive cost and quick adaptation to changes in resource, etc. Otherwise ambiguity, usage of synonymous tags and user-specific tags are the challenges that social tagging to deal with. In the view of tagging user we can find that the form of tags must rely on users'understanding and judgment. The quality of tag is determined by the user. So it is reasonable and feasible to make an obvious improvement in social tagging system by means of tagging users.In this paper, first of all, we made a deep research on social tagging system, detailed analysis of a variety of social tagging relevant concepts, system structure, advantages and problems. On this basis, we focused on the two main methods based on semantic analysis and tagging data structures in social tagging field. Then their typical algorithms were respectively discussed and analyzed. Most important, we proposed the ETB algorithm which is based on exemplary users, thus extending the algorithm's contents. After the validity and feasibility were confirmed, we built an interpersonal network model by the way of ETB algorithm.Finally in the experimental part, first of all, we used the del.icio.us data sets and took into account all factors to design the experimentation which are used to test the ETB algorithm. The experiment results show that the ETB algorithm can find the exemplary user effectively and the exemplary in multiple interests is particular effective.