A Hybrid Approach to NLP-Based Depression Detection: Integrating BERT and Word2Vec

Ganji, Yoshitha (2024) A Hybrid Approach to NLP-Based Depression Detection: Integrating BERT and Word2Vec. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (711kB) | Preview
Preview	PDF (Configuration Manual) Download (735kB) | Preview

Abstract

This study takes an innovative approach to detecting depression and suicidal ideation through social media by integrating BERT and Word2Vec into a powerful hybrid model. This model, augmented by analyses of CNN, LSTM, BiLSTM, and RoBERTa architectures, aims to substantially elevate accuracy in identifying the nuances of depressive expressions often overlooked by traditional single-model methods. Motivated by the vast potential of advanced NLP technologies, our comprehensive strategy harnesses BERT's exceptional contextual understanding and Word2Vec's in-depth semantic analysis. This dual approach addresses the complex challenge of decoding the myriad emotional cues within social media text that signal mental health concerns. This research journey, ambitious in its scope, not only pioneered the development of the hybrid model but also critically assessed its performance against other significant neural networks. This broad analysis confirmed the hybrid model's superior capability in capturing the subtleties of emotional states online, achieving a remarkable 0.90 accuracy rate. This work contributes significantly to both theoretical and practical domains. Theoretically, it opens new dialogues about the synergy between various NLP models for more accurate emotional detection, establishing new standards in the field. Practically, it signals a breakthrough in creating real-time monitoring tools for early mental health intervention on social media, offering scalable solutions that could revolutionize public health strategies

Item Type:	Thesis (Masters)
Supervisors:	Name Email Siddig, Abubakr UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > Healthcare Industry Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning R Medicine > RA Public aspects of medicine > RA790 Mental Health
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	18 Aug 2025 13:30
Last Modified:	18 Aug 2025 13:30
URI:	https://norma.ncirl.ie/id/eprint/8563

Actions (login required)

View Item