Actions
  • shareshare
  • link
  • cite
  • add
add
auto_awesome_motion View all 3 versions
Publication . Article . Conference object . 2022

Bayesian BERT for Trustful Hate Speech Detection

Miok, Kristian; Škrlj, Blaž; Zaharie, Daniela; Robnik-Šikonja, Marko;
Open Access
English
Abstract
Hate speech is an important problem in the management of user-generated content. In order to remove offensive content or ban misbehaving users, content moderators need reliable hate speech detectors. Recently, deep neural networks based on transformer architecture, such as (multilingual) BERT model, achieve superior performance in many natural language classification tasks, including hate speech detection. So far, these methods have not been able to quantify their output in terms of reliability. We propose a Bayesian method using Monte Carlo Dropout within the attention layers of the transformer models to provide well-calibrated reliability estimates. We evaluate the introduced approach on hate speech detection problems in several languages. Our approach not only improves the classification performance of the state-of-the-art multilingual BERT model but the computed reliability scores also significantly reduce the workload in inspection of offending cases and in reannotation campaigns.
Funded by
EC| EMBEDDIA
Project
EMBEDDIA
Cross-Lingual Embeddings for Less-Represented Languages in European News Media
  • Funder: European Commission (EC)
  • Project Code: 825153
  • Funding stream: H2020 | RIA
moresidebar