Text classification for suicide related tweets

Fatima Chiroma, Han Liu, Mihaela Cocea

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

57 Downloads (Pure)

Abstract

Online social networks have become a vital medium for communication. With these platforms, users have the freedom to share their opinions as well as receive information from a diverse group of people. Although this could be beneficial, there are some growing concerns regarding its negative impact on the safety of its users such as the spread of suicidal ideation. Therefore, in this study, we aim to determine the performance of machine classifiers in identifying suicide-related text from Twitter (tweets). The experiment for the study was conducted using four popular machine classifiers: Decision Tree, Naive Bayes, Random Forest and Support Vector Machine. The results of the experiment showed an F-measure ranging from 0.346 to 0.778 for suicide-related communication, with the best performance being achieved using the Decision Tree classifier.
Original languageEnglish
Title of host publicationProceedings of the 2018 International Conference on Machine Learning and Cybernetics (ICMLC)
PublisherIEEE
Pages587-592
Number of pages6
Volume2
ISBN (Electronic)978-1-5386-5214-5
ISBN (Print)978-1-5386-5215-2
DOIs
Publication statusPublished - 12 Nov 2018
Event2018 International Conference on Machine Learning and Cybernetics - http://www.icmlc.com/icmlc/welcome.html, Chengdu, China
Duration: 15 Jul 201818 Jul 2018

Publication series

NameInternational Conference on Machine Learning and Cybernetics (ICMLC)
PublisherIEEE
ISSN (Print)2160-133X
ISSN (Electronic)2160-1348

Conference

Conference2018 International Conference on Machine Learning and Cybernetics
Abbreviated titleICMLC 2018
Country/TerritoryChina
CityChengdu
Period15/07/1818/07/18

Keywords

  • Text classification
  • Machine Classifier
  • Social media
  • Suicide

Fingerprint

Dive into the research topics of 'Text classification for suicide related tweets'. Together they form a unique fingerprint.

Cite this