Naslov
Preemptive Toxic Language Detection in Wikipedia Comments Using Thread-Level Context

Autori
Karan, Mladen ; Šnajder, Jan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the Third Workshop on Abusive Language Online / - Firenza : München : Association for Computational Linguistics (ACL), 2019, 129-134

Skup
7th Workshop on Balto-Slavic Natural Language Processing. Association for Computational Linguistics

Mjesto i datum
Firenca, Italija, 28.07.2019. - 02.08.2019

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
hate speech, natural language processing
(govor mržnje, obrada prirodnog jezika)

Sažetak
We address the task of automatically detecting toxic content in user generated texts. We focus on exploring the potential for preemptive moderation, i.e., predicting whether a particular conversation thread will, in the future, incite a toxic comment. Moreover, we perform preliminary investigation of whether a model that jointly considers all comments in a conversation thread outperforms a model that considers only individual comments. Using an existing dataset of conversations among Wikipedia contributors as a starting point, we compile a new large-scale dataset for this task consisting of labeled comments and comments from their conversation threads.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA