| Summary: | 碩士 === 國立臺北科技大學 === 電機工程系研究所 === 99 === This thesis proposes some algorithms for clustering Facebook friends based on their interest. We firstly review some of existing algorithms for clustering communities in social networks, and then introduce how to retrieve data from Facebook. Finally, we propose two novel approaches, messages-similarity and friends-similarity, for clustering Facebook friends. Experimental results show that the proposed algorithms are more suitable to cluster Facebook friends than existing algorithms.
The main contributions of this thesis are: (1) Unlike the traditional clustering algorithms, which are based on the topology of social networks, the proposed algorithms are based on friends’ interest. (2) Help users understand their friends’ interest. (3) Prevent users from sharing their messages with those who are not interested.
|
|---|
