A Comparative Study of Traditional Machine Learning, Deep Learning, and Large Language Models for Mental Health Forecasting using Smartphone Sensing Data

TL;DR

This study benchmarks traditional ML, deep learning, and LLMs for mental health prediction using smartphone data, highlighting the superior performance of DL models and the benefits of personalization in forecasting severe mental health issues.

Contribution

It provides the first comprehensive comparison of ML, DL, and LLM approaches for mental health forecasting with a large longitudinal dataset, exploring various modeling strategies.

Findings

Deep learning models, especially Transformers, outperform other approaches.

Personalization significantly enhances prediction accuracy for severe mental health states.

Transformers achieved a Macro-F1 score of 0.58 in overall performance.

Abstract

Smartphone sensing offers an unobtrusive and scalable way to track daily behaviors linked to mental health, capturing changes in sleep, mobility, and phone use that often precede symptoms of stress, anxiety, or depression. While most prior studies focus on detection that responds to existing conditions, forecasting mental health enables proactive support through Just-in-Time Adaptive Interventions. In this paper, we present the first comprehensive benchmarking study comparing traditional machine learning (ML), deep learning (DL), and large language model (LLM) approaches for mental health forecasting using the College Experience Sensing (CES) dataset, the most extensive longitudinal dataset of college student mental health to date. We systematically evaluate models across temporal windows, feature granularities, personalization strategies, and class imbalance handling. Our results show that DL models, particularly Transformer (Macro-F1 = 0.58), achieve the best overall performance, while LLMs show strength in contextual reasoning but weaker temporal modeling. Personalization substantially improves forecasts of severe mental health states. By revealing how different modeling approaches interpret phone sensing behavioral data over time, this work lays the groundwork for next-generation, adaptive, and human-centered mental health technologies that can advance both research and real-world well-being.