There is a newer version of this record available.

Conference paper Open Access

Bayesian BERT for Trustful Hate Speech Detection

Miok, Kristian; Škrlj, Blaž; Zaharie, Daniela; Robnik-Šikonja, Marko


Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
  <dc:creator>Miok, Kristian</dc:creator>
  <dc:creator>Škrlj, Blaž</dc:creator>
  <dc:creator>Zaharie, Daniela</dc:creator>
  <dc:creator>Robnik-Šikonja, Marko</dc:creator>
  <dc:date>2022-03-28</dc:date>
  <dc:description>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.</dc:description>
  <dc:identifier>https://zenodo.org/record/6389324</dc:identifier>
  <dc:identifier>10.5281/zenodo.6389324</dc:identifier>
  <dc:identifier>oai:zenodo.org:6389324</dc:identifier>
  <dc:language>eng</dc:language>
  <dc:relation>info:eu-repo/grantAgreement/EC/H2020/825153/</dc:relation>
  <dc:relation>doi:10.5281/zenodo.6389323</dc:relation>
  <dc:relation>url:https://zenodo.org/communities/embeddia</dc:relation>
  <dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
  <dc:rights>https://creativecommons.org/licenses/by/4.0/legalcode</dc:rights>
  <dc:title>Bayesian BERT for Trustful Hate Speech Detection</dc:title>
  <dc:type>info:eu-repo/semantics/conferencePaper</dc:type>
  <dc:type>publication-conferencepaper</dc:type>
</oai_dc:dc>
75
58
views
downloads
All versions This version
Views 7513
Downloads 5816
Data volume 34.2 MB9.3 MB
Unique views 629
Unique downloads 5216

Share

Cite as