Research On Knowledge Topic Discovery And Application Based On Tag-clustering

Under web2.0environment, the Internet users pay more attention to the interaction between users. They are free to share resources and ideas. Everyone in the Internet has become a source of knowledge; they change from a pure knowledge consumer into producer, publisher and disseminator, content produced rich and varied, diversity in form, which is cause knowledge resources organization and management more complex. At the same time the user’s demand are diversification and socialization. People cannot get the needful knowledge from the Internet, but also the pursuit of real life continued into the network society, which requires more convenient, efficient, socialization of knowledge organization and management techniques, to share and manage the knowledge which favorite by users. And then keep it easy to others to exchange or share.The present knowledge service platform system in isolation, the data amount is limited, cannot support the realization of cross-domain data; It is difficult to deal data of different forms, different physical models and different formats; Processing level is low; The existing technology for unstructured data is not effective; The growth of information cause the difficult to meet the users knowledge needs. The traditional service mode based on expert knowledge is limited, the breadth and intensity are extremely limited, and so the need for a more effective knowledge exchanging and sharing method become urgent.The article focus on the knowledge topics discovery based on tag-clustering, according to the pluralistic need of users; make use of the method of tag-clustering to deal with data. Firstly, carried out a comprehensive survey about this field home and abroad research work, have a full understanding of advantage and weakness about the tag.Secondly,take into account the relation between users, tag and knowledge bring forth an knowledge topic discovery model based on tag-clustering to discover the collective topic focused by users, and structure a knowledge service framework. Finally, take a experiment to validate the effective, and application value of knowledge service.