David-Paul Niland

Mihael Arcan, David-Paul Niland

Mental health disorders affect over one-fifth of adults globally, yet detecting such conditions from text remains challenging due to the subtle and varied nature of symptom expression. This study evaluates multiple approaches for mental health detection, comparing Large Language Models (LLMs) such as Llama and GPT with classical machine learning and transformer-based architectures including BERT, XLNet, and Distil-RoBERTa. Using the DAIC-WOZ dataset of clinical interviews, we fine-tuned models for anxiety, depression, and stress classification and applied synthetic data generation to mitigate class imbalance. Results show that Distil-RoBERTa achieved the highest F1 score (0.883) for GAD-2, while XLNet outperformed others on PHQ tasks (F1 up to 0.891). For stress detection, a zero-shot synthetic approach (SD+Zero-Shot-Basic) reached an F1 of 0.884 and ROC AUC of 0.886. Findings demonstrate the effectiveness of transformer-based models and highlight the value of synthetic data in improving recall and generalization. However, careful calibration is required to prevent precision loss. Overall, this work emphasizes the potential of combining advanced language models and data augmentation to enhance automated mental health assessment from text.

Mihael Arcan, David-Paul Niland, Fionn Delahunty

Mental health challenges pose considerable global burdens on individuals and communities. Recent data indicates that more than 20% of adults may encounter at least one mental disorder in their lifetime. On the one hand, the advancements in large language models have facilitated diverse applications, yet a significant research gap persists in understanding and enhancing the potential of large language models within the domain of mental health. On the other hand, across various applications, an outstanding question involves the capacity of large language models to comprehend expressions of human mental health conditions in natural language. This study presents an initial evaluation of large language models in addressing this gap. Due to this, we compare the performance of Llama-2 and ChatGPT with classical Machine as well as Deep learning models. Our results on the DAIC-WOZ dataset show that transformer-based models, like BERT or XLNet, outperform the large language models.

David W. Vinson, Mihael Arcan, David-Paul Niland et al.

Employee well-being is a critical concern in the contemporary workplace, as highlighted by the American Psychological Association's 2021 report, indicating that 71% of employees experience stress or tension. This stress contributes significantly to workplace attrition and absenteeism, with 61% of attrition and 16% of sick days attributed to poor mental health. A major challenge for employers is that employees often remain unaware of their mental health issues until they reach a crisis point, resulting in limited utilization of corporate well-being benefits. This research addresses this challenge by presenting a groundbreaking stress detection algorithm that provides real-time support preemptively. Leveraging automated chatbot technology, the algorithm objectively measures mental health levels by analyzing chat conversations, offering personalized treatment suggestions in real-time based on linguistic biomarkers. The study explores the feasibility of integrating these innovations into practical learning applications within real-world contexts and introduces a chatbot-style system integrated into the broader employee experience platform. This platform, encompassing various features, aims to enhance overall employee well-being, detect stress in real time, and proactively engage with individuals to improve support effectiveness, demonstrating a 22% increase when assistance is provided early. Overall, the study emphasizes the importance of fostering a supportive workplace environment for employees' mental health.