IRMA-International.org: Creator of Knowledge
Information Resources Management Association
Advancing the Concepts & Practices of Information Resources Management in Modern Organizations

A Deep Learning Model for Textual Content Moderation Across Social Platforms Using LSTM and BERT

A Deep Learning Model for Textual Content Moderation Across Social Platforms Using LSTM and BERT
View Sample PDF
Author(s): Mohd Adnan (Interdisciplinary Centre for Artificial Intelligence, Aligarh Muslim University, India), K. Bismil (Interdisciplinary Centre for Artificial Intelligence, Aligarh Muslim University, India), Mohd Imran (Khwaja Moinuddin Chishti Language University, Lucknow, India)and Tasleem Jamal (Khwaja Moinuddin Chishti Language University, Lucknow, India)
Copyright: 2027
Pages: 28
Source title: Encyclopedia of Modern Artificial Intelligence
Source Author(s)/Editor(s): Mehdi Khosrow-Pour, D.B.A. (Founding Editor-in-Chief, Information Resources Management Journal (IRMJ), USA)
DOI: 10.4018/407629

Purchase

View A Deep Learning Model for Textual Content Moderation Across Social Platforms Using LSTM and BERT on the publisher's website for pricing and purchasing information.

Abstract

The proliferation of user-generated content on digital platforms has introduced significant challenges related to content moderation, particularly the detection of toxic, hateful, and abusive language. Consequently, artificial intelligence and NLP techniques have emerged as promising solutions for automating this task. This article investigates the use of deep learning models specifically Long Short-Term Memory (LSTM) networks and Bidirectional Encoder Representations from Transformers (BERT) for automated toxic content classification. This study utilizes both models individually and in a hybrid configuration to classify comments into three categories: safe, abusive, and hateful. The experimental setup involves pre-processing the Jigsaw Toxic Comment Classification dataset, applying tokenization, stop word removal, and balancing techniques to improve class distribution. Evaluation metrics accuracy, precision, recall, and F1-score are used to assess model performance. Results indicate that BERT significantly outperforms LSTM, achieving an F1-score of 89.6% compared to LSTM's 88.18%.

Related Content

Frederic Andres. © 2027. 14 pages.
Kalsoom Safdar, Khairul Najmy Abdul Rani, Mohd Aminudin Jamlos, Siti Julia Rosli, Muhammad Usman Younus, Zanab Safdar. © 2027. 27 pages.
Bani Adam, Binastya Anggara Sekti, Muhammad Adi Zacky Zahran. © 2027. 24 pages.
Swetha Margaret T. A., Renuka Devi D.. © 2027. 31 pages.
Maurice Saluschke, Michael Schulz. © 2027. 30 pages.
Mirjam Sepesy Maučec, Gregor Donaj. © 2027. 16 pages.
Jorge A. Ruiz-Vanoye, Ocotlan Diaz-Parra, Ricardo A. Barrera-Cámara, Alejandro Fuentes-Penna, Francisco R. Trejo-Macotela, Jaime Aguilar-Ortiz, Eric Simancas-Acevedo. © 2027. 21 pages.
Body Bottom