A Study of Multilingual Toxic Text Detection Approaches under Imbalanced Sample Distribution

A Study of Multilingual Toxic Text Detection Approaches under Imbalanced Sample Distribution

Multilingual characteristics, lack of annotated data, and imbalanced sample distribution are the three main challenges for toxic comment analysis in a multilingual setting. This paper proposes a multilingual toxic text classifier which adopts a novel fusion strategy that combines different loss func...

Full description

Bibliographic Details
Main Authors: Guizhe Song, Degen Huang, Zhifeng Xiao
Format: Article
Language:English
Published: MDPI AG 2021-05-01
Series:Information
Subjects:
toxic comment
imbalanced positive and negative samples
pre-training models
multilingual classification
XLM-RoBERTa
MBERT
Online Access:https://www.mdpi.com/2078-2489/12/5/205

Internet

https://www.mdpi.com/2078-2489/12/5/205

Similar Items