Books CONTRIBUTED AND ACKNOWLEDGED
Explainable AI: An Introduction to Interpretable XAI
Explainable AI: An Introduction to Interpretable XAI
Explainable AI: An Introduction to Interpretable XAI
Transformers for Machine Learning: A Deep Drive
Explainable AI: An Introduction to Interpretable XAI
Explainable AI: An Introduction to Interpretable XAI
Research Publication
PsychBERT: A Mental Health Language Model for Social Media Mental Health Behavioral Analysis
Abstract—Mental health behaviors are now recognized as
primary factors contributing to suicide. Social media text is an
increasingly important modality for detecting mental behaviors.
Currently, there is no taxonomy and no comprehensive dataset
that machine learning researchers can employ as a benchmark
to evaluate and advance research to address this problem.
Fragmented efforts in the community also demonstrate that
it remains challenging to recognize text relevant to mental
health analysis and distinguish behaviors, such as depression
and social anxiety. This paper puts forth a novel mental health
language model that addresses these challenges. The paper
makes several contributions. First, it proposes a taxonomy and
puts forth a comprehensive dataset of social media text for
the community. Second, it proposes a two-stage framework,
first discriminating text relevant to mental health from nonrelevant
text and then carrying out multi-class classification
for detection of mental health behaviors. Third, it proposes
a novel mental health language model, PsychBERT, which is
pretrained on a large corpus of biomedical literature on mental
health, as well as social media data. Fourth, the proposed
framework additionally incorporates components that enhance
its explainability. Our evaluation shows that our proposed
framework, strongly leveraging PsychBERT, is both effective,
outperforming state-of-the-art methods, and interpretable. The
taxonomy, dataset, and pre-trained PsychBERT model are made
publicly available. The pretrained language model is available at
