A News Event Detection Algorithm Based on Key Elements Recognition
Xiaoting Qu, Juan Yang, Bin Wu, Haiming Xin · 2016
With the development of the Internet continues accelerating, network news has gradually become an indispensable source where people get news events. How to obtain the core news events from network news and grasp the dynamic information of the society has gradually become one of the problems that people concerned. This paper researched the features of network news and proposed an event detection algorithm based on key elements recognition. The algorithm based on single-pass algorithm and inherited the briefness principle from it. In addition, the algorithm is combined with key elements recognition, which meets the requirements of instantaneity and accuracy in event detection. We adopted Named Entity Recognition and Part-of-Speech tagging in characteristic words selection, combined Vector Space Model with temporal property in vectorization, and imported the vector space of events to improve the quality of clustering. Experimental results proved the effectiveness and practicability of our algorithm.