Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation

Van Cong Pham; Thair Al-Dala’in

doi:10.1007/978-3-031-87647-9_10

Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation

Van Cong Pham, Thair Al-Dala’in

Western Sydney University

Western Sydney University

Research output: Chapter in Book / Conference Paper › Chapter › peer-review

Abstract

This study emphasizes the importance of natural language processing (NLP) in identifying hate speech on social media when it comes to Vietnamese language by synthesizing many previous papers. A finding from the study is that models like Bidirectional Encoder Representations from Transformers (BERT) and its variations have outperformed Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) in terms of accuracy and reliability. By studying previous research papers on Vietnamese language, it is found that with the complexity and diversity of Vietnamese language such as homonyms, polysemy, slang, regional languages and so on, they are considered as major linguistic challenges that prevent the success in detecting hate speech using NLP tools. However, there is a substantial research gap existing in this field well as in the tools used to process and filter them. To enhance the capacity to identify and reduce harmful speech and support the development of healthier online communities, future research should concentrate on comprehending and resolving these obstacles.

Original language	English
Title of host publication	Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25)
Editors	Kevin Daimi, Abeer Al Sadoon
Place of Publication	Switzerland
Publisher	Springer
Pages	113-126
Number of pages	14
ISBN (Electronic)	9783031876479
ISBN (Print)	9783031876462
DOIs	https://doi.org/10.1007/978-3-031-87647-9_10
Publication status	Published - 2025
Event	International Conference on Advances in Computing Research - Radisson Hotel Nice Airport, Nice, France Duration: 7 Jul 2025 → 8 Jul 2025 Conference number: 3rd

Publication series

Name	Lecture Notes in Networks and Systems
Volume	1346
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	International Conference on Advances in Computing Research
Abbreviated title	ACR
Country/Territory	France
City	Nice
Period	7/07/25 → 8/07/25

Keywords

Bidirectional Encoder Representations from Transformers (BERT)
Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM)
Hate Speech (HS)
Natural language processing (NLP)
Vietnamese

Access to Document

10.1007/978-3-031-87647-9_10

Cite this

Pham, V. C., & Al-Dala’in, T. (2025). Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation. In K. Daimi, & A. Al Sadoon (Eds.), Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25) (pp. 113-126). (Lecture Notes in Networks and Systems; Vol. 1346). Springer. https://doi.org/10.1007/978-3-031-87647-9_10

Pham, Van Cong ; Al-Dala’in, Thair. / Natural language processing (NLP) for hate speech detection in Vietnamese language : challenges and implementation. Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25). editor / Kevin Daimi ; Abeer Al Sadoon. Switzerland : Springer, 2025. pp. 113-126 (Lecture Notes in Networks and Systems).

@inbook{b8726f29f0914f1b9fd151d8cbc10c0a,

title = "Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation",

abstract = "This study emphasizes the importance of natural language processing (NLP) in identifying hate speech on social media when it comes to Vietnamese language by synthesizing many previous papers. A finding from the study is that models like Bidirectional Encoder Representations from Transformers (BERT) and its variations have outperformed Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) in terms of accuracy and reliability. By studying previous research papers on Vietnamese language, it is found that with the complexity and diversity of Vietnamese language such as homonyms, polysemy, slang, regional languages and so on, they are considered as major linguistic challenges that prevent the success in detecting hate speech using NLP tools. However, there is a substantial research gap existing in this field well as in the tools used to process and filter them. To enhance the capacity to identify and reduce harmful speech and support the development of healthier online communities, future research should concentrate on comprehending and resolving these obstacles.",

keywords = "Bidirectional Encoder Representations from Transformers (BERT), Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), Hate Speech (HS), Natural language processing (NLP), Vietnamese",

author = "Pham, {Van Cong} and Thair Al-Dala{\textquoteright}in",

year = "2025",

doi = "10.1007/978-3-031-87647-9_10",

language = "English",

isbn = "9783031876462",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer",

pages = "113--126",

editor = "Kevin Daimi and {Al Sadoon}, Abeer",

booktitle = "Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25)",

note = "International Conference on Advances in Computing Research, ACR ; Conference date: 07-07-2025 Through 08-07-2025",

}

Pham, VC & Al-Dala’in, T 2025, Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation. in K Daimi & A Al Sadoon (eds), Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25). Lecture Notes in Networks and Systems, vol. 1346, Springer, Switzerland, pp. 113-126, International Conference on Advances in Computing Research, Nice, France, 7/07/25. https://doi.org/10.1007/978-3-031-87647-9_10

Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation. / Pham, Van Cong; Al-Dala’in, Thair.
Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25). ed. / Kevin Daimi; Abeer Al Sadoon. Switzerland: Springer, 2025. p. 113-126 (Lecture Notes in Networks and Systems; Vol. 1346).

Research output: Chapter in Book / Conference Paper › Chapter › peer-review

TY - CHAP

T1 - Natural language processing (NLP) for hate speech detection in Vietnamese language

T2 - International Conference on Advances in Computing Research

AU - Pham, Van Cong

AU - Al-Dala’in, Thair

N1 - Conference code: 3rd

PY - 2025

Y1 - 2025

N2 - This study emphasizes the importance of natural language processing (NLP) in identifying hate speech on social media when it comes to Vietnamese language by synthesizing many previous papers. A finding from the study is that models like Bidirectional Encoder Representations from Transformers (BERT) and its variations have outperformed Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) in terms of accuracy and reliability. By studying previous research papers on Vietnamese language, it is found that with the complexity and diversity of Vietnamese language such as homonyms, polysemy, slang, regional languages and so on, they are considered as major linguistic challenges that prevent the success in detecting hate speech using NLP tools. However, there is a substantial research gap existing in this field well as in the tools used to process and filter them. To enhance the capacity to identify and reduce harmful speech and support the development of healthier online communities, future research should concentrate on comprehending and resolving these obstacles.

AB - This study emphasizes the importance of natural language processing (NLP) in identifying hate speech on social media when it comes to Vietnamese language by synthesizing many previous papers. A finding from the study is that models like Bidirectional Encoder Representations from Transformers (BERT) and its variations have outperformed Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) in terms of accuracy and reliability. By studying previous research papers on Vietnamese language, it is found that with the complexity and diversity of Vietnamese language such as homonyms, polysemy, slang, regional languages and so on, they are considered as major linguistic challenges that prevent the success in detecting hate speech using NLP tools. However, there is a substantial research gap existing in this field well as in the tools used to process and filter them. To enhance the capacity to identify and reduce harmful speech and support the development of healthier online communities, future research should concentrate on comprehending and resolving these obstacles.

KW - Bidirectional Encoder Representations from Transformers (BERT)

KW - Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM)

KW - Hate Speech (HS)

KW - Natural language processing (NLP)

KW - Vietnamese

UR - http://www.scopus.com/inward/record.url?scp=105003623035&partnerID=8YFLogxK

UR - https://go.openathens.net/redirector/westernsydney.edu.au?url=https://doi.org/10.1007/978-3-031-87647-9_10

U2 - 10.1007/978-3-031-87647-9_10

DO - 10.1007/978-3-031-87647-9_10

M3 - Chapter

AN - SCOPUS:105003623035

SN - 9783031876462

T3 - Lecture Notes in Networks and Systems

SP - 113

EP - 126

BT - Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25)

A2 - Daimi, Kevin

A2 - Al Sadoon, Abeer

PB - Springer

CY - Switzerland

Y2 - 7 July 2025 through 8 July 2025

ER -

Pham VC, Al-Dala’in T. Natural language processing (NLP) for hate speech detection in Vietnamese language: challenges and implementation. In Daimi K, Al Sadoon A, editors, Proceedings of the 3rd International Conference on Advances in Computing Research (ACR'25). Switzerland: Springer. 2025. p. 113-126. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-87647-9_10