Bursty Topic Detection on Twitter

摘要：

Twitter is a user-generated content system that allows its users to share short text messages, called tweets. Every day millions of users post an extraordinarily large number of small messages, it is almost impossible for people to read all of them. In this paper, we propose a novel topic detection technique that permits to retrieve in real-time the burstiest topics expressed by the community. It first identifies bursty features by time series analysis over the twitter stream. The computation of weight of features takes account of various parameters including position and the number of replies. Then these features are grouped into topics by unsupervised learning clustering algorithm. A preliminary evaluation is carried out on an implementation of this technique that shows promising results.

关键词： Bursty topic detection Twitter analysis Time series analysis

作者: YanYan Du

作者单位: School of computer University of Wuhan Wuhan, Hubei Province, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 174-177

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）