An attention-based hybrid architecture with explainability for depressive social media text detection in Bangla

Ghosh, T., Al Banna, M.H., Al Nahian, M.J., Uddin, M.N., Kaiser, M.S. and Mahmud, M. ORCID: 0000-0002-2037-8348, 2023. An attention-based hybrid architecture with explainability for depressive social media text detection in Bangla. Expert Systems with Applications, 213 (Part C): 119007. ISSN 0957-4174

[img]
Preview
Text
1618450_Mahmud.pdf - Published version

Download (3MB) | Preview

Abstract

Mental health has become a major concern in recent years. Social media have been increasingly used as platforms to gain insight into a person’s mental health condition by analysing the posts and comments, which are textual in nature. By analysing these texts, depressive posts can be detected. To facilitate this process, this work presents an attention-based bidirectional Long Short-Term Memory (LSTM)- Convolutional Neural Network (CNN) based model to detect depressive Bangla social media texts, which is lighter and more robust than the conventional models and provides better performance. A dataset containing such Bangla texts was also developed in this work to mitigate the scarcity. Different preprocessing stages were followed, and three embeddings were used in this task. Thanks to the attention mechanism, the proposed model achieved an accuracy of 94.3% with 92.63% of sensitivity and 95.12% of specificity. When tested on other languages, such as English, the proposed model performed remarkably. The robustness and explainability of the proposed model were also discussed in this paper. Additionally, when compared with classical machine learning models, ensemble approaches, transformers, other similar models, and existing architectures, the proposed model outperformed them.

Item Type: Journal article
Publication Title: Expert Systems with Applications
Creators: Ghosh, T., Al Banna, M.H., Al Nahian, M.J., Uddin, M.N., Kaiser, M.S. and Mahmud, M.
Publisher: Elsevier BV
Date: 1 March 2023
Volume: 213
Number: Part C
ISSN: 0957-4174
Identifiers:
NumberType
10.1016/j.eswa.2022.119007DOI
1618450Other
Rights: © 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND licence (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Divisions: Schools > School of Science and Technology
Record created by: Laura Ward
Date Added: 15 Nov 2022 10:06
Last Modified: 15 Nov 2022 10:06
URI: https://irep.ntu.ac.uk/id/eprint/47398

Actions (login required)

Edit View Edit View

Views

Views per month over past year

Downloads

Downloads per month over past year