MOPI-HFRS: A Multi-objective Personalized Health-aware Food Recommendation System with LLM-enhanced Interpretation

TL;DR

This work tackles health-aware food recommendation by balancing user preferences, personalized healthiness, and nutritional diversity while delivering interpretable reasoning. It introduces two NHANES-derived Health and Nutrition Recommendation Bipartite Graph benchmarks and the MOPI-HFRS framework, which fuses health-aware graph structure learning, Pareto multi-objective optimization, and knowledge-infused LLM reasoning. Empirical results show MOPI-HFRS outperforms state-of-the-art baselines on multi-objective metrics and delivers higher-quality explanations, aided by two domain-aware prompting strategies. The approach holds practical impact for personalized, health-promoting, and explainable dietary recommendations at scale, with avenues for extending personalization and interpretability in health nutrition systems.

Abstract

The prevalence of unhealthy eating habits has become an increasingly concerning issue in the United States. However, major food recommendation platforms (e.g., Yelp) continue to prioritize users' dietary preferences over the healthiness of their choices. Although efforts have been made to develop health-aware food recommendation systems, the personalization of such systems based on users' specific health conditions remains under-explored. In addition, few research focus on the interpretability of these systems, which hinders users from assessing the reliability of recommendations and impedes the practical deployment of these systems. In response to this gap, we first establish two large-scale personalized health-aware food recommendation benchmarks at the first attempt. We then develop a novel framework, Multi-Objective Personalized Interpretable Health-aware Food Recommendation System (MOPI-HFRS), which provides food recommendations by jointly optimizing the three objectives: user preference, personalized healthiness and nutritional diversity, along with an large language model (LLM)-enhanced reasoning module to promote healthy dietary knowledge through the interpretation of recommended results. Specifically, this holistic graph learning framework first utilizes two structure learning and a structure pooling modules to leverage both descriptive features and health data. Then it employs Pareto optimization to achieve designed multi-facet objectives. Finally, to further promote the healthy dietary knowledge and awareness, we exploit an LLM by utilizing knowledge-infusion, prompting the LLMs with knowledge obtained from the recommendation model for interpretation.

Figures (6)

• Figure 1: The illustration of the impact of a poor diet, and the comprehensive and challenging nature of the multi-objective personalized interpretable health-aware food recommendation task.
• Figure 2: The schema of Health and Nutrition Recommendation Bipartite Graph.
• Figure 3: (a) The overall framework of MOPI-HFRS, which consists of three major components: (b) a health-aware structure learning module with feature-based structure learning, and (c) healthy edge structure learning; (d) a Pareto optimization module and a knowledge-infused reasoning module.
• Figure 4: The prompts (highlighted in blue and red) generated from the MOPI-HFRS for interpreting the healthiness of recommendation results.
• Figure 5: Performance of the reasoning task. Results are reported as mean and std. of BERT Scores and BLEU Scores.

Theorems & Definitions (3)

• definition 1: Bipartite Graph for Recommendation
• definition 2: Signed Bipartite Graph
• definition 3: Pareto Optimality