Information credibility: A probabilistic graphical model for identifying credible influenza posts on social media

Qiaozhen Guo, Wei (Wayne) Huang, Kai Huang, Xiao Liu

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

2 Scopus citations

Abstract

Social media is an important data source to compliment traditional epidemic surveillance. However, misinformation in social media hinders the exploitation of valuable information. Analysis of information credibility has drawn much attention of academia in recent years. In this paper, we focus on analyzing the credibility of influenza posts published on Sina Weibo. We propose a semi-supervised probabilistic graphical model to jointly learn the interactions between user trustworthiness, content reliability, and post credibility. To test the performance of the approach, we apply it to identify credible influenza posts published from May 2013 to June 2014 on Sina Weibo. Random Forests and the Bayesian Network are used as baselines for evaluation. The results show that our approach performs effectively with the highest average accuracy of 71.7%, f-measure 51%. Our proposed framework significantly outperformed the baselines in detecting credible influenza posts on Sina Weibo.

Original languageEnglish (US)
Title of host publicationSmart Health - International Conference, ICSH 2015, Revised Selected Papers
EditorsHsinchun Chen, Daniel Dajun Zeng, Xiaolong Zheng, Scott J. Leischow
PublisherSpringer Verlag
Pages131-142
Number of pages12
ISBN (Print)9783319291741
DOIs
StatePublished - 2016
EventInternational Conference for Smart Health, ICSH 2015 - Phoenix, United States
Duration: Nov 17 2015Nov 18 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9545
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference for Smart Health, ICSH 2015
Country/TerritoryUnited States
CityPhoenix
Period11/17/1511/18/15

Keywords

  • Credibility
  • Influenza
  • Objectivity
  • Probabilistic graphical models
  • Reliability
  • Sina weibo
  • Trustworthiness

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Information credibility: A probabilistic graphical model for identifying credible influenza posts on social media'. Together they form a unique fingerprint.

Cite this