Hot Events Detection for Chinese Microblogs Based on the TH-LDA Model

DOI

10.2991/tlicsc-18.2018.26How to use a DOI?

Keywords

Hot Events Detection; LDA Model; Chinese Microblogs.

Abstract

Nowadays, many unexpected topics in the society initiate on the microblog platform and spread rapidly, and some of them finally become hot events. The technology for detecting these hot events on the microblog platform, has exerted a very positive influence on the discovery of the latest social hotspots and the timely perception of the internet public opinion. This paper analyzed the studies for detecting microblog hot events, and disclosed the existing methods may omit the untagged microblogs and thus lead to the failure to detect the subevents. To address this issue, we combined Hashtag and Time with the topic model of LDA, and proposed the TH-LDA model to improve the effectiveness of hot events detection for Chinese microblogs. Experiments on microblogs datasets demonstrate that the proposed TH-LDA model can effectively obtain untagged microblogs, and then realize the subevents detection of hot events with a high accuracy.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jiahui Chen
AU  - Qingxia Shang
AU  - Hailing Xiong
PY  - 2018/12
DA  - 2018/12
TI  - Hot Events Detection for Chinese Microblogs Based on the TH-LDA Model
BT  - Proceedings of the 2018 International Conference on Transportation & Logistics, Information & Communication, Smart City (TLICSC 2018)
PB  - Atlantis Press
SP  - 157
EP  - 166
SN  - 1951-6851
UR  - https://doi.org/10.2991/tlicsc-18.2018.26
DO  - 10.2991/tlicsc-18.2018.26
ID  - Chen2018/12
ER  -