Tailored motivational message generation: A model and practical framework for real-time physical activity coaching.

Area of Science:

• Computer Science
• Behavioral Science
• Human-Computer Interaction

Background:

• Behavior change applications often use motivational messages.
• Current message generation methods may lack personalization and real-time adaptation.
• Tailoring messages involves considering time, intention, content, and presentation.

Purpose of the Study:

• To present a comprehensive framework for automatic, real-time generation of tailored motivational messages.
• To extend existing ontology-based approaches for message generation in coaching systems.
• To provide practical examples of implementing tailoring rules for user engagement.

Main Methods:

• A linear modular system for message generation.
• Incorporation of user and context properties into message tailoring modules.
• Analysis of existing literature and a corpus of motivational messages.
• Development of tailoring rules for linguistic presentation.

Main Results:

• A practical framework for real-time, tailored message generation is presented.
• The framework accounts for user and context properties in message tailoring.
• Examples demonstrate the implementation of tailoring rules, focusing on physical activity promotion.

Conclusions:

• The proposed framework offers a structured approach to personalized message generation in behavior change applications.
• The model extends current methods by integrating user and context-specific tailoring.
• Future research should focus on framework application, evaluation of tailoring components, and user/context model improvement.