並列摘要
In the new era of Web 2.0, Microblogging has become a new way to let users communicate with friends promptly and efficiently. Due to the huge amount messages generated in microblogging sites, it is an important task to find users and messages that you are interesting. We observed that users in Plurk surfer from the inconvenient and insufficient interaction interface. The recommended friends on Plurk are not easy to find. Users can not also easily find the topics in that they get interested to follow. Therefore, in this paper we aim to provide an easy mechanism for users to find other users they are interesting by clustering users with their discussion topics. We link the topics that are mentioned in the same time and are from the same theme. These topics will be considered a topic event. We call this method ''Timing and Correlation Clustering'' (TCC). We can then build a complete event list and recommend to users that are interested in this event and let users know this event and other users who are also interested in this event.
被引用紀錄
延伸閱讀
- 黃千籙（2014）。Clustering News Articles Based on Social Behavior〔碩士論文，國立臺北科技大學〕。華藝線上圖書館。https://doi.org/10.6841/NTUT.2014.00086
- 洪秉儒（2011）。Web Text Clustering with Dynamic Theme〔碩士論文，國立中央大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0031-1903201314411137
- 許經國（2014）。The Analysis and Improvements of People Clustering in Video〔碩士論文，國立交通大學〕。華藝線上圖書館。https://doi.org/10.6842/NCTU.2014.00895
- 黃威穎（2006）。Hierarchically Dynamic Clustering of Web Search Results〔碩士論文，元智大學〕。華藝線上圖書館。https://doi.org/10.6838/YZU.2006.00124
- Lai, Y. M. (2011). Automatic identification of hot topics and user clusters from online discussion forums [master's thesis, The University of Hong Kong]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0029-1812201200020454