Skip to Main content Skip to Navigation
Conference papers

Generalisability of Topic Models in Cross-corpora Abusive Language Detection

Tulika Bose 1 Irina Illina 1 Dominique Fohr 1
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Rapidly changing social media content calls for robust and generalisable abuse detection models. However, the state-of-the-art supervised models display degraded performance when they are evaluated on abusive comments that differ from the training corpus. We investigate if the performance of supervised models for cross-corpora abuse detection can be improved by incorporating additional information from topic models, as the latter can infer the latent topic mixtures from unseen samples. In particular, we combine topical information with representations from a model tuned for classifying abusive comments. Our performance analysis reveals that topic models are able to capture abuse-related topics that can transfer across corpora, and result in improved generalisability.
Document type :
Conference papers
Complete list of metadata

https://hal.inria.fr/hal-03212196
Contributor : Tulika Bose <>
Submitted on : Thursday, April 29, 2021 - 2:01:07 PM
Last modification on : Friday, April 30, 2021 - 4:13:34 PM

File

14_Paper-5.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03212196, version 1

Citation

Tulika Bose, Irina Illina, Dominique Fohr. Generalisability of Topic Models in Cross-corpora Abusive Language Detection. 2021 Workshop on NLP4IF: Censorship, Disinformation, and Propaganda, NAACL, Jun 2021, Mexico city/Virtual, Mexico. ⟨hal-03212196⟩

Share

Metrics

Record views

21

Files downloads

35