Research on Hot Topic Discovery Based on Intelligent Campus Information Service Platform
Fengling Wang · 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · 2019
This paper analyzes the common text clustering algorithm and text representation model for the hot topics found in the campus. Through the use of the word segmentation system, the text word segmentation and keyword extraction are proposed, and a knowledge representation model of the message text is proposed. The study of vector space model uses the word frequency of statistical messages to determine k initial clustering centers and improve the optimization algorithm k-Means. The detailed process and concrete steps of the hot topic discovery algorithm are given, and the topic discovery module is verified. The experimental results show that the intelligent topic clustering process can be processed faster and better, and the clustering accuracy can be kept high when the number of reported messages is large.