Threat detection in online discussions

Wester, Aksel Ladegård; Øvrelid, Lilja; Velldal, Erik; Hammer, Hugo Lewi
Chapter, Peer reviewed, Chapter
Accepted version
Thumbnail
View/Open
URI
https://hdl.handle.net/10642/4287
Date
2016
Metadata
Show full item record
Collections
Original version
Wester WA, Øvrelid L, Velldal E, Hammer HL: Threat detection in online discussions. In: Balahur A, van der Goot E, Vossen P, Montoyo A. Proceedings of the 7th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis WASSA@NAACL-HLT 2016 (WASSA 2016), 2016. Association for Computational Linguistics p. 66-71  
Abstract
This paper investigates the effect of various

types of linguistic features (lexical, syntactic

and semantic) for training classifiers to detect

threats of violence in a corpus of YouTube

comments. Our results show that combina-

tions of lexical features outperform the use of

more complex syntactic and semantic features

for this task.