MLtoGAI: Semantic Web based with Machine Learning for Enhanced Disease Prediction and Personalized Recommendations using Generative AI

Explainability is central to responsible healthcare AI: The system prioritizes making AI recommendations understandable to patients by using ChatGPT to generate clear, personalized explanations rather than presenting opaque predictions, addressing a key concern in responsible AI deployment in medical settings.

Structured knowledge improves both accuracy and trustworthiness: By building a reusable disease ontology (a formal knowledge structure), the system achieves "substantial advancements in accuracy" while making the reasoning behind predictions more transparent and auditable—important for clinical validation and user trust.

Combining multiple AI approaches can enhance responsible outcomes: The integration of machine learning for pattern recognition with semantic web technology and rule-based systems creates a system that is simultaneously more accurate and more explainable, suggesting that responsible AI isn't about choosing one technology but thoughtfully combining complementary approaches.

User comprehension is a measurable success metric: The paper emphasizes "user satisfaction" and "user comprehension" as key outcomes alongside prediction accuracy, reflecting a responsible AI principle that healthcare systems must be designed for the people who use them, not just for technical performance.

Personalization with accountability: The system generates individualized health advice while maintaining the ability to trace how recommendations were derived through semantic rules and structured knowledge, balancing the benefits of personalization with the need for explainability and accountability in healthcare decisions.