A Hybrid RoBERTa-Nvidia Nemotron-4 Framework for Unsupervised and Fine-Tuned Analysis of Mental Health Discourse on Reddit
Mental Health, Natural Language Processing, RoBERTa, nvidia nemotron-4, Unsupervised Training, Supervised Training, Reddit Data, Suicide, Depression, Motivation, Happiness, Personalized Remedies, Mental Health Classification, AI Intervention
Mental health issues, including suicide ideation,
sadness, and apathy, are prevalent concerns that necessitate
innovative solutions. Social media platforms, especially Reddit,
offer a unique chance to evaluate user-generated data to derive
insights about mental health. This work employs the RoBERTa
base model for unsupervised training on a dataset including
3,408 entries from Reddit. The research concentrates on subjects
including suicide, happiness, despair, and motivation. To achieve
a high degree of precision in the classification of mental health
conditions, the model undergoes further fine-tuning, facilitated
by a supervised dataset including 50,000 classified records.
The integration of the nvidia nemotron-4 model within the
framework enables dynamic user input and personalised query
results. Besides classification, the system provides actionable
remedies tailored to the user’s identified mental condition.
This enables individuals to undertake preventive measures. This
hybrid method demonstrates the promise of AI-driven solutions
for mental health analysis. This method provides a robust and
engaging tool for early intervention and assistance.
"A Hybrid RoBERTa-Nvidia Nemotron-4 Framework for Unsupervised and Fine-Tuned Analysis of Mental Health Discourse on Reddit", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.9, Issue 12, page no.a157-a165, December-2024, Available :https://ijsdr.org/papers/IJSDR2412020.pdf
Volume 9
Issue 12,
December-2024
Pages : a157-a165
Paper Reg. ID: IJSDR_300009
Published Paper Id: IJSDR2412020
Downloads: 000347159
Research Area: Science and Technology
Country: Visakhapatnam, Andhra Pradesh, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
