Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies

  • 11 Citations

Abstract

Cyberbullying is becoming a major concern in online environments with troubling consequences. However, most of the technical studies have focused on the detection of cyberbullying through identifying harassing comments rather than preventing the incidents by detecting the bullies. In this work we study the automatic detection of bully users on YouTube. We compare three types of automatic detection: an expert system, supervised machine learning models, and a hybrid type combining the two. All these systems assign a score indicating the level of “bulliness‿ of online bullies. We demonstrate that the expert system outperforms the machine learning models. The hybrid classifier shows an even better performance.
Original languageUndefined
Title of host publicationProceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014
Place of PublicationBerlin
PublisherSpringer Verlag
Pages275-281
Number of pages7
ISBN (Print)978-3-319-06482-6
DOIs
StatePublished - May 2014

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Verlag
Volume8436
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Expert systems
Learning systems
Classifiers

Keywords

  • bulliness score
  • EWI-24725
  • IR-91449
  • METIS-305877
  • Cyberbullying
  • Identity identification
  • Expert system

Cite this

Dadvar, M., Trieschnigg, R. B., & de Jong, F. M. G. (2014). Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies. In Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014 (pp. 275-281). (Lecture Notes in Computer Science; Vol. 8436). Berlin: Springer Verlag. DOI: 10.1007/978-3-319-06483-3_25

Dadvar, M.; Trieschnigg, Rudolf Berend; de Jong, Franciska M.G. / Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies.

Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014. Berlin : Springer Verlag, 2014. p. 275-281 (Lecture Notes in Computer Science; Vol. 8436).

Research output: Scientific - peer-reviewConference contribution

@inbook{cc6a32718a9041eca2b073fda290f008,
title = "Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies",
abstract = "Cyberbullying is becoming a major concern in online environments with troubling consequences. However, most of the technical studies have focused on the detection of cyberbullying through identifying harassing comments rather than preventing the incidents by detecting the bullies. In this work we study the automatic detection of bully users on YouTube. We compare three types of automatic detection: an expert system, supervised machine learning models, and a hybrid type combining the two. All these systems assign a score indicating the level of “bulliness‿ of online bullies. We demonstrate that the expert system outperforms the machine learning models. The hybrid classifier shows an even better performance.",
keywords = "bulliness score, EWI-24725, IR-91449, METIS-305877, Cyberbullying, Identity identification, Expert system",
author = "M. Dadvar and Trieschnigg, {Rudolf Berend} and {de Jong}, {Franciska M.G.}",
note = "10.1007/978-3-319-06483-3_25",
year = "2014",
month = "5",
doi = "10.1007/978-3-319-06483-3_25",
isbn = "978-3-319-06482-6",
series = "Lecture Notes in Computer Science",
publisher = "Springer Verlag",
pages = "275--281",
booktitle = "Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014",

}

Dadvar, M, Trieschnigg, RB & de Jong, FMG 2014, Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies. in Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014. Lecture Notes in Computer Science, vol. 8436, Springer Verlag, Berlin, pp. 275-281. DOI: 10.1007/978-3-319-06483-3_25

Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies. / Dadvar, M.; Trieschnigg, Rudolf Berend; de Jong, Franciska M.G.

Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014. Berlin : Springer Verlag, 2014. p. 275-281 (Lecture Notes in Computer Science; Vol. 8436).

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies

AU - Dadvar,M.

AU - Trieschnigg,Rudolf Berend

AU - de Jong,Franciska M.G.

N1 - 10.1007/978-3-319-06483-3_25

PY - 2014/5

Y1 - 2014/5

N2 - Cyberbullying is becoming a major concern in online environments with troubling consequences. However, most of the technical studies have focused on the detection of cyberbullying through identifying harassing comments rather than preventing the incidents by detecting the bullies. In this work we study the automatic detection of bully users on YouTube. We compare three types of automatic detection: an expert system, supervised machine learning models, and a hybrid type combining the two. All these systems assign a score indicating the level of “bulliness‿ of online bullies. We demonstrate that the expert system outperforms the machine learning models. The hybrid classifier shows an even better performance.

AB - Cyberbullying is becoming a major concern in online environments with troubling consequences. However, most of the technical studies have focused on the detection of cyberbullying through identifying harassing comments rather than preventing the incidents by detecting the bullies. In this work we study the automatic detection of bully users on YouTube. We compare three types of automatic detection: an expert system, supervised machine learning models, and a hybrid type combining the two. All these systems assign a score indicating the level of “bulliness‿ of online bullies. We demonstrate that the expert system outperforms the machine learning models. The hybrid classifier shows an even better performance.

KW - bulliness score

KW - EWI-24725

KW - IR-91449

KW - METIS-305877

KW - Cyberbullying

KW - Identity identification

KW - Expert system

U2 - 10.1007/978-3-319-06483-3_25

DO - 10.1007/978-3-319-06483-3_25

M3 - Conference contribution

SN - 978-3-319-06482-6

T3 - Lecture Notes in Computer Science

SP - 275

EP - 281

BT - Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014

PB - Springer Verlag

ER -

Dadvar M, Trieschnigg RB, de Jong FMG. Experts and Machines against Bullies: A Hybrid Approach to Detect Cyberbullies. In Proceedings of the 27th Canadian Conference on Artificial Intelligence, Canadian AI 2014. Berlin: Springer Verlag. 2014. p. 275-281. (Lecture Notes in Computer Science). Available from, DOI: 10.1007/978-3-319-06483-3_25