Explainable Detection of Depression Status Shifts from User Digital Traces

TL;DR

This paper presents an explainable framework that detects and analyzes depression status shifts over time using user digital traces, combining multiple models for interpretability and improved change point detection.

Contribution

It introduces a novel, interpretable approach that integrates BERT-based signals and large language models to identify and describe mental health transitions from social media data.

Findings

The approach outperforms direct LLM reporting in coherence and informativeness.

It achieves higher coverage of user history and better sensitivity to change points.

Ablation studies confirm the importance of temporal modeling and segmentation.

Abstract

Every day, users generate digital traces (e.g., social media posts, chats, and online interactions) that are inherently timestamped and may reflect aspects of their mental state. These traces can be organized into temporal trajectories that capture how a user's mental health signals evolve, including phases of improvement, deterioration, or stability. In this work, we propose an explainable framework for detecting and analyzing depression-related status shifts in user digital traces. The approach combines multiple BERT-based models to extract complementary signals across different dimensions (e.g., sentiment, emotion, and depression severity). Such signals are then aggregated over time to construct user-level trajectories that are analyzed to identify meaningful change points. To enhance interpretability, the framework integrates a large language model to generate concise and human-readable reports that describe the evolution of mental-health signals and highlight key transitions. We evaluate the framework on two social media datasets. Results show that the approach produces more coherent and informative summaries than direct LLM-based reporting, achieving higher coverage of user history, stronger temporal coherence, and improved sensitivity to change points. An ablation study confirms the contribution of each component, particularly temporal modeling and segmentation. Overall, the method provides an interpretable view of mental health signals over time, supporting research and decision making without aiming at clinical diagnosis.