The Science and Information Organization

Online social networks have seen a rapid growth inrecent years. A key aspect of many of such networks is that theyare rich in content and social interactions. Users of socialnetworks connect with each other and forming their owncommunities. With the evolution of huge communities hosted bysuch websites, users suffer from managing information overloadand it is become hard to extract useful information. Thus, usersneed a mechanism to filter online social streams they receive aswell as enable them to interact with most similar users. In thispaper, we address the problem of personalizing dissemination ofrelevant information in knowledge sharing social network. Theproposed framework identifies the most appropriate user(s) toreceive specific post by calculating similarity between target userand others. Similarity between users within OSN is calculatedbased on users’ social activity which is an integration of contentpublished as well as social pattern Application of this frameworkto a representative subset of a large real-world social network:the user/community network of the blog service stack overflow isillustrated here. Experiments show that the proposed modeloutperform tradition similarity methods. Keywords—social network; content similarity measurement;Information retrieval; Information dissemination