A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community

With the rapid development of digital technologies, data-driven methods have demonstrated commendable performance in the toxic text detection task. However, several challenges remain unresolved, including the inability to fully capture the nuanced semantic information embedded in text languages, the...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenting Fan, Haoyan Song, Jun Zhang
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/13/2136
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850115829634433024
author Wenting Fan
Haoyan Song
Jun Zhang
author_facet Wenting Fan
Haoyan Song
Jun Zhang
author_sort Wenting Fan
collection DOAJ
description With the rapid development of digital technologies, data-driven methods have demonstrated commendable performance in the toxic text detection task. However, several challenges remain unresolved, including the inability to fully capture the nuanced semantic information embedded in text languages, the lack of robust mechanisms to handle the inherent uncertainty of text languages, and the utilization of static fusion strategies for multi-view information. To address these issues, this paper proposes a comprehensive and dynamic toxic text detection method. Specifically, we design a multi-view feature augmentation module by combining bidirectional long short-term memory and BERT as a dual-stream framework. This module captures a more holistic representation of semantic information by learning both local and global features of texts. Next, we introduce an entropy-oriented invariant learning module by minimizing the conditional entropy between view-specific representations to align consistent information, thereby enhancing the representation generalization. Meanwhile, we devise a trustworthy text recognition module by defining the Dirichlet function to model uncertainty estimation of text prediction. And then, we perform the evidence-based information fusion strategy to dynamically aggregate decision information between views with the help of the Dirichlet distribution. Through these components, the proposed method aims to overcome the limitations of traditional methods and provide a more accurate and reliable solution for toxic language detection. Finally, extensive experiments on the two real-world datasets show the effectiveness and superiority of the proposed method in comparison with seven methods.
format Article
id doaj-art-437f6223b40643dea04ead3e5edd5b6b
institution OA Journals
issn 2227-7390
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-437f6223b40643dea04ead3e5edd5b6b2025-08-20T02:36:28ZengMDPI AGMathematics2227-73902025-06-011313213610.3390/math13132136A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese CommunityWenting Fan0Haoyan Song1Jun Zhang2School of European Language and Culture Studies, Dalian University of Foreign Languages, Dalian 116044, ChinaUniversity International College, Macau University of Science and Technology, Macau 999078, ChinaGraduate School of Education, Dalian University of Technology, Dalian 116024, ChinaWith the rapid development of digital technologies, data-driven methods have demonstrated commendable performance in the toxic text detection task. However, several challenges remain unresolved, including the inability to fully capture the nuanced semantic information embedded in text languages, the lack of robust mechanisms to handle the inherent uncertainty of text languages, and the utilization of static fusion strategies for multi-view information. To address these issues, this paper proposes a comprehensive and dynamic toxic text detection method. Specifically, we design a multi-view feature augmentation module by combining bidirectional long short-term memory and BERT as a dual-stream framework. This module captures a more holistic representation of semantic information by learning both local and global features of texts. Next, we introduce an entropy-oriented invariant learning module by minimizing the conditional entropy between view-specific representations to align consistent information, thereby enhancing the representation generalization. Meanwhile, we devise a trustworthy text recognition module by defining the Dirichlet function to model uncertainty estimation of text prediction. And then, we perform the evidence-based information fusion strategy to dynamically aggregate decision information between views with the help of the Dirichlet distribution. Through these components, the proposed method aims to overcome the limitations of traditional methods and provide a more accurate and reliable solution for toxic language detection. Finally, extensive experiments on the two real-world datasets show the effectiveness and superiority of the proposed method in comparison with seven methods.https://www.mdpi.com/2227-7390/13/13/2136toxic language detectioninvariant representation learningtrustworthy-driven adaptive fusion
spellingShingle Wenting Fan
Haoyan Song
Jun Zhang
A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
Mathematics
toxic language detection
invariant representation learning
trustworthy-driven adaptive fusion
title A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
title_full A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
title_fullStr A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
title_full_unstemmed A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
title_short A Novel Trustworthy Toxic Text Detection Method with Entropy-Oriented Invariant Representation Learning for Portuguese Community
title_sort novel trustworthy toxic text detection method with entropy oriented invariant representation learning for portuguese community
topic toxic language detection
invariant representation learning
trustworthy-driven adaptive fusion
url https://www.mdpi.com/2227-7390/13/13/2136
work_keys_str_mv AT wentingfan anoveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity
AT haoyansong anoveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity
AT junzhang anoveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity
AT wentingfan noveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity
AT haoyansong noveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity
AT junzhang noveltrustworthytoxictextdetectionmethodwithentropyorientedinvariantrepresentationlearningforportuguesecommunity