Abstract
Microblogging, as a popular social media service, has become a new information channel for users to receive and exchange the most up-to-date information on current events. Consequently, it is crucial to detect new emerging events and to discover the key posts which have the potential to actively disseminate the events through microblogging. However, traditional event detection models require human intervention for detecting the number of topics to be explored, which significantly reduces the efficiency and accuracy of event detection. Most of the existing methods focus only on event detection and are unable to discover the key posts related to the events, making it challenging to track momentous events in timely manner. To tackle these problems, a HITS (Hypertext Induced Topic Search) based topic decision method, named TD-HITS is proposed. TD-HITS can automatically detect the number of topics as well as discovering associated key posts from a large number of posts. The experimental results are based on a Twitter dataset to demonstrate the effectiveness of our proposed methods for both detecting events and identifying key posts.