The best way to conference proceedings by Francis Academic Press

Web of Proceedings - Francis Academic Press
Web of Proceedings - Francis Academic Press

Research on Multi-view Clustering Data Mining Algorithms Based on Microblog

Download as PDF

DOI: 10.25236/dpaic.2018.006

Author(s)

Bigeng Zheng, Xinrui Chen

Corresponding Author

Bigeng Zheng

Abstract

Micro-blog is a new type of social network platform emerging in recent years. With the development of the Internet and the continuous popularization of mobile personal terminals, the number of users and content of microblog have achieved a leap growth. By applying clustering analysis to weibo users, some user groups with common characteristics can be obtained. Analyzing these groups can result in topics that are of common interest to users within the group. In this paper, using the LDA theme model that can extract text production model of implied theme, on micro blog text subject extraction, text on the hidden topic space vector distribution, with the traditional text clustering to the quantitative model to combine, thereby effectively solving the problem of data sparseness and semantic loss, implemented by using the method of multi-level clustering topic found that topic and use the keyword extraction method is presented.Simulation results demonstrate the superiority of the proposed algorithm.

Keywords

Micro Blog, Text Data Mining, Eigenvectors, Similarity Matrix