Skip to main navigation Skip to search Skip to main content

Automatic discovery of abusive thai language usages in social networks

  • Mahidol University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

18 Citations (Scopus)

Abstract

Social networks have become a standard means of communication that allows a massive amount of users to interact and consume information anywhere and anytime. In Thailand, millions of users have access to social networks, a majority of which include young children. The colloquial nature of social media inherently encourages certain expressions of language that do not conform to the standard, some of which may be considered abusive and offensive. Such ill-mannered language fashion has become increasingly used by a large number of Thai social media users. If these abusive languages are exposed to adolescents without proper guidance, they could compulsorily develop a familiar attitude towards such language styles. To address the issue, we present a set of algorithms based on machine learning, that automatically detect abusive Thai language in social networks. Our best results yield 86% f-measure (88.73% precision and 83.53% recall).

Original languageEnglish
Title of host publicationDigital Libraries
Subtitle of host publicationData, Information, and Knowledge for Digital Lives - 19th International Conference on Asia-Pacific Digital Libraries, ICADL 2017, Proceedings
EditorsSally Jo Cunningham, Songphan Choemprayong, Fabio Crestani
PublisherSpringer Verlag
Pages267-278
Number of pages12
ISBN (Print)9783319702315
DOIs
Publication statusPublished - 2017
Event19th International Conference on Asia-Pacific Digital Libraries, ICADL 2017 - Bangkok, Thailand
Duration: 13 Nov 201715 Nov 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10647 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference19th International Conference on Asia-Pacific Digital Libraries, ICADL 2017
Country/TerritoryThailand
CityBangkok
Period13/11/1715/11/17

Keywords

  • Abusive language detection
  • Large scale social networks
  • Thai natural language processing

Fingerprint

Dive into the research topics of 'Automatic discovery of abusive thai language usages in social networks'. Together they form a unique fingerprint.

Cite this