Food Recommendation With Balancing Comfort and Curiosity
Yuto Sakai, Qiang Ma
TL;DR
This paper tackles balancing comfort (familiar tastes) and curiosity (local novelty) in travel-food recommendations. It introduces two quantitative scoring schemes, Kernel Density Scoring ($KDS$) and Mahalanobis Distance Scoring ($MDS$), to assess how candidate foods fit a user’s past history, and a ranking mechanism that combines taste and ingredients to compute a balance score ($Score^{total}_{f}$). A crowdsourced dataset with foods and explicit comfort/curiosity judgments is constructed and used to evaluate the methods; $Score_f$ is derived as $-\log p(f)$ in $KDS$ and as $d_{out}/d_{in}$ in $MDS$, with $Score^{total}$ used to rank items by comfort-curiosity trade-off. Experimental results show the $MDS$ method generally outperforms a random baseline in ROC-AUC and significantly improves when mapping curiosity to taste and comfort to ingredients, demonstrating practical potential for travel-context food recommendations.
Abstract
Food is a key pleasure of traveling, but travelers face a trade-off between exploring curious new local food and choosing comfortable, familiar options. This creates demand for personalized recommendation systems that balance these competing factors. To the best of our knowledge, conventional recommendation methods cannot provide recommendations that offer both curiosity and comfort for food unknown to the user at a travel destination. In this study, we propose new quantitative methods for estimating comfort and curiosity: Kernel Density Scoring (KDS) and Mahalanobis Distance Scoring (MDS). KDS probabilistically estimates food history distribution using kernel density estimation, while MDS uses Mahalanobis distances between foods. These methods score food based on how their representation vectors fit the estimated distributions. We also propose a ranking method measuring the balance between comfort and curiosity based on taste and ingredients. This balance is defined as curiosity (return) gained per unit of comfort (risk) in choosing a food. For evaluation the proposed method, we newly collected a dataset containing user surveys on Japanese food and assessments of foreign food regarding comfort and curiosity. Comparing our methods against the existing method, the Wilcoxon signed-rank test showed that when estimating comfort from taste and curiosity from ingredients, the MDS-based method outperformed the Baseline, while the KDS-based method showed no significant differences. When estimating curiosity from taste and comfort from ingredients, both methods outperformed the Baseline. The MDS-based method consistently outperformed KDS in ROC-AUC values.