Potential cyberbullying detection in social media platforms based on a multi-task learning framework
The proliferation of online violence has given rise to a spate of malignant incidents, necessitating a renewed focus on the identification of cyberbullying comments. Text classification lies at the heart of efforts to tackle this pernicious problem. The identification of cyberbullying comments prese...
Saved in:
Main Authors: | , |
---|---|
Format: | Article |
Published: |
Growing Science
2024
|
Subjects: | |
Online Access: | http://eprints.um.edu.my/44737/ https://doi.org/10.5267/j.ijdns.2023.10.021 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.um.eprints.44737 |
---|---|
record_format |
eprints |
spelling |
my.um.eprints.447372024-07-11T00:54:44Z http://eprints.um.edu.my/44737/ Potential cyberbullying detection in social media platforms based on a multi-task learning framework Xingyi, Guo Adnan, Hamedi Mohd HN Social history and conditions. Social problems. Social reform The proliferation of online violence has given rise to a spate of malignant incidents, necessitating a renewed focus on the identification of cyberbullying comments. Text classification lies at the heart of efforts to tackle this pernicious problem. The identification of cyberbullying comments presents unique challenges that call for innovative solutions. In contrast to traditional text classification tasks, cyberbullying comments are often accompanied by subtle and arbitrary expressions that can confound even the most sophisticated classification networks, resulting in low recognition accuracy and effectiveness. To address this challenge, a novel approach is proposed that leverages the BERT pre-training model for word embedding to retain the hidden semantic information in the text. Building on this foundation, the BiSRU++ model which combines attentional mecha-nisms is used to further extract contextual features of comments. A multi-task learning framework is employed for joint training of sentiment analysis and cyberbullying detection to improve the model's classification accuracy and generalization ability. The proposed model is no longer en-tirely reliant on a sensitive word dictionary, and experimental results demonstrate its ability to better understand semantic information compared to traditional models, facilitating the identification of potential online cyberbullying comments. © 2024 by the authors; licensee Growing Science, Canada. Growing Science 2024 Article PeerReviewed Xingyi, Guo and Adnan, Hamedi Mohd (2024) Potential cyberbullying detection in social media platforms based on a multi-task learning framework. International Journal of Data and Network Science, 8 (1). pp. 25-34. ISSN 2561-8148, DOI https://doi.org/10.5267/j.ijdns.2023.10.021 <https://doi.org/10.5267/j.ijdns.2023.10.021>. https://doi.org/10.5267/j.ijdns.2023.10.021 10.5267/j.ijdns.2023.10.021 |
institution |
Universiti Malaya |
building |
UM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Malaya |
content_source |
UM Research Repository |
url_provider |
http://eprints.um.edu.my/ |
topic |
HN Social history and conditions. Social problems. Social reform |
spellingShingle |
HN Social history and conditions. Social problems. Social reform Xingyi, Guo Adnan, Hamedi Mohd Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
description |
The proliferation of online violence has given rise to a spate of malignant incidents, necessitating a renewed focus on the identification of cyberbullying comments. Text classification lies at the heart of efforts to tackle this pernicious problem. The identification of cyberbullying comments presents unique challenges that call for innovative solutions. In contrast to traditional text classification tasks, cyberbullying comments are often accompanied by subtle and arbitrary expressions that can confound even the most sophisticated classification networks, resulting in low recognition accuracy and effectiveness. To address this challenge, a novel approach is proposed that leverages the BERT pre-training model for word embedding to retain the hidden semantic information in the text. Building on this foundation, the BiSRU++ model which combines attentional mecha-nisms is used to further extract contextual features of comments. A multi-task learning framework is employed for joint training of sentiment analysis and cyberbullying detection to improve the model's classification accuracy and generalization ability. The proposed model is no longer en-tirely reliant on a sensitive word dictionary, and experimental results demonstrate its ability to better understand semantic information compared to traditional models, facilitating the identification of potential online cyberbullying comments. © 2024 by the authors; licensee Growing Science, Canada. |
format |
Article |
author |
Xingyi, Guo Adnan, Hamedi Mohd |
author_facet |
Xingyi, Guo Adnan, Hamedi Mohd |
author_sort |
Xingyi, Guo |
title |
Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
title_short |
Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
title_full |
Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
title_fullStr |
Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
title_full_unstemmed |
Potential cyberbullying detection in social media platforms based on a multi-task learning framework |
title_sort |
potential cyberbullying detection in social media platforms based on a multi-task learning framework |
publisher |
Growing Science |
publishDate |
2024 |
url |
http://eprints.um.edu.my/44737/ https://doi.org/10.5267/j.ijdns.2023.10.021 |
_version_ |
1805881163851497472 |
score |
13.18916 |