Advancing Depression Detection in Social Media: A Multimodal Aspect-Level Sentiment Analysis Approach

DOI

10.2991/978-94-6463-512-6_44How to use a DOI?

Keywords

Depression Detection; Multimodal Aspect-Based Sentiment Analysis; Deep Learning

Abstract

Depression, a prevalent mental illness, increasingly manifests in the digital expressions of individuals on social media platforms. This paper explores the use of multimodal aspect-level sentiment analysis for detecting depressive tendencies from social media data, a method that surpasses traditional unimodal approaches in granularity and adaptability. By integrating textual and visual cues from users’ posts, this paper’s approach employs the Target-Oriented Multimodal Bidirectional Encoder Representations from Transformers (TOM-BERT) framework. This deep learning model is fine-tuned to discern subtle indicators of depression by analyzing the interplay between different types of data inputs. This paper’s experimental setup compares this method against conventional models primarily focused on single-mode data analysis, demonstrating its superior capability in identifying depressive signals. Results reveal that this paper’s multimodal approach not only captures a richer spectrum of emotional expressions but also enhances the accuracy of depression detection. This research underscores the potential of advanced sentiment analysis techniques in mental health monitoring, particularly in leveraging the nuanced data available through social networks.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yuxin Liu
PY  - 2024
DA  - 2024/09/23
TI  - Advancing Depression Detection in Social Media: A Multimodal Aspect-Level Sentiment Analysis Approach
BT  - Proceedings of the 2024 International Conference on Artificial Intelligence and Communication (ICAIC 2024)
PB  - Atlantis Press
SP  - 409
EP  - 420
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-512-6_44
DO  - 10.2991/978-94-6463-512-6_44
ID  - Liu2024
ER  -