Fine-Tuned DistilBERT and Interactive Dashboard for Early Detection of Suicidal Ideation in Social Media Posts

Marquez Salazar, Diego (2025) Fine-Tuned DistilBERT and Interactive Dashboard for Early Detection of Suicidal Ideation in Social Media Posts. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (847kB) | Preview
Preview	PDF (Configuration Manual) Download (411kB) | Preview

Abstract

The early detection of suicidal ideation in social media content can save lives by enabling timely interventions. In this work, we present the development of a fine-tuned DistilBERT model on a pre-processed dataset to classify texts as indicative of suicide risk or not. The model was integrated into an interactive dashboard built with Streamlit, which allows both manual input of individual texts and bulk analysis through CSV file uploads. The tool displays classification results along with estimated probabilities and provides additional visualizations, such as word frequency distributions, to help uncover patterns in the data. Experimental evaluation using 5-fold cross-validation demonstrated outstanding performance, achieving accuracy above 93% and ROC-AUC scores exceeding 0.90 This solution offers a practical and socially impactful contribution, aimed at mental health professionals, researchers, and organizations seeking to monitor and prevent suicide related risks within online communities.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Haque, Rejwanul UNSPECIFIED
Uncontrolled Keywords:	Suicide prevention; DistilBERT; NLP; sentiment analysis; Streamlit; dashboard; mental health
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing R Medicine > RA Public aspects of medicine > RA790 Mental Health Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet > World Wide Web > Websites > Online social networks T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet > World Wide Web > Websites > Online social networks
Divisions:	School of Computing > Master of Science in Artificial Intelligence for Business
Depositing User:	Ciara O'Brien
Date Deposited:	24 Aug 2026 10:55
Last Modified:	24 Aug 2026 10:55
URI:	https://norma.ncirl.ie/id/eprint/9591

Actions (login required)

View Item