| With the continuous development of SNS, microblogging services have already become a necessary part of daily life of many Chinese people. Microblogging is a broadcast social network platform that people share brief instant messages via a one-way follow mechanism. Being a social medium, its features of short text, real time, sociality and media greatly reduce the time interval between release and spread of information, which favours rapid spread of news. At present, microblogging has become a main aggregation and outbreak site of online public opinions. Effectively detecting emerging topics from microblogging platforms has great practical significance for common users, online businessmen and government departments.This thesis sums up the existing research findings in the field of emerging topics, and then designs and implements an emerging topic detection system based on Sina Weibo, which is called ETD. The system collects data of users and tweets of Sina Weibo in real time, improves integrity and consistence of the data set via a certain scheduling strategy. It utilizes a novel emerging topic detection model, which can detect emerging topics from mass tweets accurately. Finally it displays information of the detected topics in Web front-end by adopting some latest data visualization technologies.First, this thesis introduces the related theories and technical background of emerging topic detection. Second, requirement analysis and system design of the emerging topic detection system based on Chinese microblogging are depicted in detail. Third, the detailed design and implementation of each subsystem are highlighted, including data acquisition, emerging topic detection and visualization. Then unit and integration testing are adopted, which indicates that the entire system achieves the prospective design goals. In the end, the thesis summarizes the whole work of the system, looks forward to future work, and summarizes the work and achievements of the author during the post-graduate period. |