Predicting mass incidents from weibo

WenWen Li, Yang Zhou, Tingting Lu, Tingshao Zhu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The outbreak of mass incidents severely affects the stability of society. If we can predict mass incidents in advance, we may find the solution to avoid the confliction in time. Some of the existing approaches rely on emotional modeling. Much research has been conducted on microblog incident detection using statistical models, like LASSO regression method, Dynamic Query Expansion (DQE) and so on. In this paper, we propose to combine sentiment analysis and statistical methods, and uses LASSO regression method for mass incidents prediction. Experiments on Qingdao demonstrated that our proposed approach achieves a good performance.

Original languageEnglish (US)
Title of host publicationHuman Centered Computing - 2nd International Conference, HCC 2016, Revised Selected Papers
PublisherSpringer Verlag
Pages895-900
Number of pages6
Volume9567
ISBN (Print)9783319318530
DOIs
StatePublished - 2016
Externally publishedYes
Event2nd International Conference on Human Centered Computing, HCC 2016 - Colombo, Sri Lanka
Duration: Jan 7 2016Jan 9 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9567
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd International Conference on Human Centered Computing, HCC 2016
CountrySri Lanka
CityColombo
Period1/7/161/9/16

Keywords

  • Dynamic query expansion
  • Event forecasting
  • LASSO
  • Mass incidents
  • Sentiment analysis

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Predicting mass incidents from weibo'. Together they form a unique fingerprint.

Cite this