An AI-Based Framework for Interpretable Mental Health Literacy Segmentation and Decision Support.

Area of Science:

• Mental Health Research
• Educational Psychology
• Health Informatics

Background:

• Mental Health Literacy (MHL) is crucial for identifying mental health issues and seeking help.
• Existing interventions often lack personalization, overlooking modifiable factors.
• A gap exists in tailored strategies to enhance MHL within academic settings.

Purpose of the Study:

• To develop a decision support framework assessing MHL dimensions.
• To segment academic communities based on modifiable MHL predictors.
• To inform targeted educational interventions for improved mental healthcare.

Main Methods:

• Cross-sectional survey of 385 academic participants.
• Latent Profile Analysis to identify distinct MHL profiles.
• Comparison of decision tree classifiers (C5.0, CART, CIT) for rule generation.

Main Results:

• Five distinct latent profiles of MHL were identified.
• Decision tree models provided interpretable rules for each profile.
• An interactive dashboard visualized MHL profile interactions with demographics.

Conclusions:

• The framework enables precise intervention design for academic communities.
• Actionable insights support policymakers and health professionals in resource allocation.
• Enhances institutional mental healthcare through targeted strategies and tracking.