Scalable and Explainable Diet Recommendations via Answer Set Programming

We present a logic-based approach to personalized dietary recommendations using Answer Set Programming (ASP). Our system represents user profiles, including dietary preferences, allergies, and nutritional needs, and selects compatible recipes that satisfy logical constraints while maximizing adherence to health goals. Through ASP, we ensure explainable reasoning and the ability to flexibly encode dietary rules. We illustrate the system’s behavior through two representative case studies, involving complex user requirements and food restrictions. Furthermore, we conduct a scalability analysis by generating large sets of synthetic users and recipes, evaluating the system’s performance under increasing data sizes. The results show that our ASP-based method remains interpretable and tractable for moderate-scale applications, with further optimization planned. This work contributes toward trustworthy and adaptable diet recommendation systems, paving the way for integration into personalized digital health solutions.