A cuisine-adaptive explainable AI framework for restaurant-specific meal recommendation: a Thai restaurant case study

Information from the abstract

Abstract Food recommendation systems often struggle to model culturally grounded meal composition behaviors, particularly in multi-dish dining environments where sensory balance and dish complementarity influence menu selection. This study proposes an explainable AI framework that integrates association rule mining with tri-dimensional sensory profiling to analyze dish pairing behavior in restaurant transactions. Using real-world Thai restaurant data, dishes were categorized into 15 culinary groups and profiled according to dominant taste, texture, and aroma characteristics. The Apriori algorithm was then applied to identify statistically significant co-occurrence patterns and generate interpretable association rules. By classifying dish categories and tagging them with dominant taste, texture, and aroma profiles, we apply the Apriori algorithm to generate interpretable association rules. These rules are further enriched through tri-dimensional sensory analysis to reveal cultural and sensory rationales behind dish pairings. The results demonstrate consistent patterns of contrast and harmony in texture, taste, and aroma, reflecting deeply embedded culinary logic. We formalize these insights into a rule-based recommendation system that models authentic Thai meal compositions. The framework not only advances computational food pairing research but also offers practical implications for menu design, inventory planning, and personalized food recommendation systems in culturally rich dining environments.

Why this record is monitored

This record has an Impact Signal of 85/100 based on recency, source, collaboration, and bibliographic signals. It prioritizes monitoring and is not a judgment of research quality.

Related topics: Recommender Systems and Techniques · Data Mining Algorithms and Applications · Culinary Culture and Tourism