CoChat
Sign Up
Cited by 16Recommender Systems and TechniquesInnovation in Digital Healthcare SystemsInformation Retrieval and Search Behavior

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

Open access
Zhang, Zheyuan, Wang, Zehong, Ma, Tianyi, Taneja, Varun Sameer, Nelson, Sofia, Le, Nhi Ha Lan, Murugesan, Keerthiram, Ju, Mingxuan +3 more·2025·Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1
DOI 10.1145/3690624.3709382
Open this paper in CoChat — free account

Abstract

The prevalence of unhealthy eating habits has become a growing concern in the United States. However, popular food recommendation platforms, such as Yelp, tend to prioritize users' dietary preferences over the healthiness of their choices. While some efforts have focused on developing health-aware food recommendation systems, personalization based on specific health conditions remains underexplored. Additionally, the lack of interpretability in these systems prevents users from evaluating the reliability of recommendations, limiting their practical adoption. To address these issues, we introduce two large-scale personalized health-aware food recommendation benchmarks at the first attempt. Building on this, we propose a novel framework called the Multi-Objective Personalized Interpretable Health-aware Food Recommendation System (MOPI-HFRS). This system generates food recommendations by jointly optimizing three objectives: user preference, personalized healthiness, and nutritional diversity. It also incorporates a reasoning module enhanced by large language models (LLMs) to provide interpretable recommendations that promote healthy dietary knowledge. The framework integrates descriptive features and health data using two structure learning and pooling modules within a graph learning framework. Pareto optimization is applied to balance the multi-faceted objectives. To further enhance healthy dietary knowledge, the system leverages LLMs by infusing knowledge from the recommendation model, generating meaningful interpretations for the recommendations. Extensive experiments on the proposed benchmarks demonstrate that MOPI-HFRS outperforms state-of-the-art methods by delivering diverse, healthy food recommendations alongside reliable explanations.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3690624.3709382

Cites

2 of 2 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Web search enables agents to ground answers in retrieved evidence and thereby mitigate hallucinations.

    “A representative and widely adopted form of tool use is web search [17, 19, 27], which enables agents to ground answers in retrieved evidence and thereby mitigate hallucinations [31, 33].”

    From CiteAudit: You Cited It, But Did You Read It? A Benchmark for Verifying Scientific References in the LLM Era (2026) · cites this paper as [33]

    The Case Study (p. 9) says prompt learning and information from neighboring food nodes can mitigate hallucination in food recommendations. Related Work (p. 3) mentions retrieval-augmented generation, but neither passage says web search grounds agents’ answers in retrieved evidence or mitigates hallucinations. Searches for web search, retrieved evidence, and retrieval-augmented grounding found no such passage. (LLM, conf=0.75)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 38

  1. [4][4] Andrew Blair-Stanek, Nils Holzenberger, and Benjamin Van Durme. 2023. Can GPT-3 perform statutory reasoning? arXiv (2023). -
  2. [5][5] Felix Bölz, Diana Nurbakova, Sylvie Calabretto, Armin Gerl, Lionel Brunie, and Harald Kosch. 2023. HUMMUS: A Linked, Healthiness-Aware, User-centered and Argument-Enabling Recipe Data Set for Recommendation. In RecSys . -
  3. [6][6] Xuheng Cai, Chao Huang, Lianghao Xia, and Xubin Ren. 2023. LightGCL: Simple yet effective graph contrastive learning for recommendation. ICLR (2023). -
  4. [7][7] Dorwin Cartwright and Frank Harary. 1956. Structural balance: a generalization of Heider's theory. Psychological review 63, 5 (1956), 277. -
  5. [9][9] Hongxu Chen, Yicong Li, Xiangguo Sun, Guandong Xu, and Hongzhi Yin. 2021. Temporal meta-path guided explainable recommendation. In WSDM . -
  6. [12][12] Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In NAACL . -
  7. [16][16] Mouzhi Ge, Francesco Ricci, and David Massimo. 2015. Health-aware food recommender system. In RecSys . -
  8. [18][18] Will Hamilton, Zhitao Ying, and Jure Leskovec. 2017. Inductive representation learning on large graphs. In NeurIPS . -
  9. [20][20] Xiangnan He, Tao Chen, Min-Yen Kan, and Xiao Chen. 2015. Trirank: Reviewaware explainable recommendation by modeling aspects. In CIKM . -
  10. [21][21] Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and powering graph convolution network for recommendation. In SIGIR . -
  11. [22][22] Scott Howell and Richard Kones. 2017. 'Calories in, calories out' and macronutrient intake: the hope, hype, and science of calories. American Journal of PhysiologyEndocrinology and Metabolism (2017). -
  12. [23][23] Mohamed Manzour Hussien, Angie Nataly Melo, Augusto Luis Ballardini, Carlota Salinas Maldonado, Rubén Izquierdo, and Miguel Ángel Sotelo. 2024. RAGbased Explainable Prediction of Road Users Behaviors for Automated Driving using Knowledge Graphs and Large Language Models. arXiv (2024). -
  13. [24][24] Martin Jaggi. 2013. Revisiting Frank-Wolfe: Projection-free sparse convex optimization. In International conference on machine learning . -
  14. [25][25] Mingxuan Ju, Tong Zhao, Qianlong Wen, Wenhao Yu, Neil Shah, Yanfang Ye, and Chuxu Zhang. 2023. Multi-task self-supervised graph neural networks enable stronger task generalization. ICLR (2023). -
  15. [26][26] Thomas N Kipf and Max Welling. 2017. Semi-supervised classification with graph convolutional networks. In ICLR . -
  16. [27][27] Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, et al. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. NIPS (2020). , , -
  17. [28][28] Lei Li, Yongfeng Zhang, and Li Chen. 2023. Personalized prompt learning for explainable recommendation. ACM Transactions on Information Systems (2023). -
  18. [29][29] Ming Li, Lin Li, Xiaohui Tao, and Ning Zhong. 2023. User-Meal Interaction Learning for Meal Recommendation: A Reproducibility Study. In SIGIR-AP . -
  19. [31][31] Qiyao Ma, Xubin Ren, and Chao Huang. 2024. XRec: Large Language Models for Explainable Recommendation. arXiv (2024). -
  20. [32][32] Javier Marın, Aritro Biswas, Ferda Ofli, Nicholas Hynes, Amaia Salvador, Yusuf Aytar, Ingmar Weber, and Antonio Torralba. 2021. Recipe1m+: A dataset for learning cross-modal embeddings for cooking recipes and food images. IEEE Transactions on Pattern Analysis and Machine Intelligence (2021). -
  21. [33][33] Costas Mavromatis and George Karypis. 2024. GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning. arXiv (2024). -
  22. [37][37] Yoon-Joo Park and Alexander Tuzhilin. 2008. The long tail of recommender systems and how to leverage it. In Proceedings of the 2008 ACM conference on Recommender systems . -
  23. [38][38] Matt Post. 2018. A call for clarity in reporting BLEU scores. ACL (2018). -
  24. [39][39] Amaia Salvador, Nicholas Hynes, Yusuf Aytar, Javier Marin, Ferda Ofli, Ingmar Weber, and Antonio Torralba. 2017. Learning cross-modal embeddings for cooking recipes and food images. In CVPR . -
  25. [40][40] Jialiang Shi, Takahiro Komamizu, Keisuke Doman, Haruya Kyutoku, and Ichiro Ide. 2023. RecipeMeta: Metapath-enhanced Recipe Recommendation on Heterogeneous Recipe Network. In Proceedings of the 5th ACM International Conference on Multimedia in Asia . -
  26. [42][42] Andrew Smyth, Martin J O'Donnell, Salim Yusuf, Catherine M Clase, Koon K Teo, Michelle Canavan, Donal N Reddan, and Johannes FE Mann. 2014. Sodium intake and renal outcomes: a systematic review. American journal of hypertension (2014). -
  27. [43][43] Ruixiang Tang, Xiaotian Han, Xiaoqian Jiang, and Xia Hu. 2023. Does synthetic data generation of llms help clinical text mining? arXiv (2023). -
  28. [44][44] Yijun Tian, Chuxu Zhang, Zhichun Guo, Chao Huang, Ronald Metoyer, and Nitesh V Chawla. 2022. RecipeRec: A heterogeneous graph learning model for recipe recommendation. IJCAI (2022). -
  29. [45][45] Christoph Trattner and David Elsweiler. 2017. Food recommender systems: important contributions, challenges and future research directions. arXiv preprint arXiv:1711.02760 (2017). - · arXiv
  30. [46][46] Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2018. Graph attention networks. In ICLR . -
  31. [47][47] Hongwei Wang, Fuzheng Zhang, Jialin Wang, Miao Zhao, Wenjie Li, Xing Xie, and Minyi Guo. 2018. Ripplenet: Propagating user preferences on the knowledge graph for recommender systems. In CIKM . -
  32. [49][49] Xinfeng Wang, Jin Cui, Yoshimi Suzuki, and Fumiyo Fukumoto. 2024. RDRec: Rationale Distillation for LLM-based Recommendation. arXiv (2024). -
  33. [50][50] Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. 2019. Neural graph collaborative filtering. In SIGIR . -
  34. [51][51] Jiancan Wu, Xiang Wang, Fuli Feng, Xiangnan He, Liang Chen, Jianxun Lian, and Xing Xie. 2021. Self-supervised graph learning for recommendation. In SIGIR . -
  35. [53][53] Mengyuan Yang, Mengying Zhu, Yan Wang, Linxun Chen, Yilei Zhao, Xiuyuan Wang, Bing Han, Xiaolin Zheng, and Jianwei Yin. 2024. Fine-Tuning Large Language Model Based Explainable Recommendation with Explainable Quality Reward. In AAAI . -
  36. [54][54] Hongzhi Yin, Bin Cui, Jing Li, Junjie Yao, and Chen Chen. 2012. Challenging the long tail recommendation. arXiv (2012). -
  37. [55][55] Junliang Yu, Hongzhi Yin, Xin Xia, Tong Chen, Lizhen Cui, and Quoc Viet Hung Nguyen. 2022. Are graph augmentations necessary? simple graph contrastive learning for recommendation. In SIGIR . -
  38. [56][56] Wenbin Yue, Zidong Wang, Jieyu Zhang, and Xiaohui Liu. 2021. An overview of recommendation techniques and their applications in healthcare. IEEE/CAA Journal of Automatica Sinica (2021). -

Cited by

Sorted by citations

Authors
Zhang, Z., Wang, Z., Ma, T., Taneja, V. S., Nelson, S., Le, N. H. L., Murugesan, K., Ju, M., Chawla, N. V., Zhang, C., & Ye, Y.
Affiliations
  • University of Notre Dame (US)
  • Brandeis University (US)
  • IBM (United States) (US)
  • University of Connecticut (US)
ORCID
Venue
Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1
Publisher
ACM
Published
2025-07-20
Pages
2860-2871
DOI
10.1145/3690624.3709382
OpenAlex
W4409149653
Cited by
16
References
2
Citations by year
2026: 9 · 2025: 7
FWCI
18.30
Type
conference paper
Language
en
Open access
Licence
Locations
  • ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • arXiv (Cornell University) (cc-by)
Funders
  • NSF (National Science Foundation) (IIS-2321504, IIS-2334193, IIS-2340346, IIS-2217239, CNS-2426514, CNS-2203261, and CMMI-2146076)
  • ND-IBM Tech Ethics Lab Collaborative Project
Topics
  • Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
  • Health Sciences › Health Professions › Health Information Management › Innovation in Digital Healthcare Systems
  • Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
Keywords
personalized recommendation, multi-objective optimization, interpretability, large language models, dietary diversity, food preferences, graph neural networks, structure learning, pooling, healthy eating

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review