Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{199578,
        author = {Krunal Parab and Aman Pandey and Sarthak Bahal and Vikas Singh},
        title = {AI-Powered Personalized Diet Recommendation System: An Integrated Research Study},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14104-14109},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199578},
        abstract = {Building a personalized nutrition system involves a major technical tradeoff. You have to balance strict medical dietary rules with what users actually want to eat. In this paper, we detail the software architecture and initial prototype for a new diet recommendation system called "NutriMind." We started by combining standard health calculations (like BMI and BMR) with a hybrid machine learning engine. This engine uses K-Nearest Neighbors (KNN) alongside Deep Neural Collaborative Filtering (NCF). We also wanted to fix the common hallucination issues seen when using Large Language Models (LLMs) for medical advice. To do this, we built a Retrieval-Augmented Generation (RAG) pipeline. Our RAG setup forces the LLM to output strict JSON data boundaries pulled directly from the ADA 2026 Standards of Care and DASH diet protocols. For the data layer, we integrated the 2.23-million record RecipeNLG dataset and applied Min-Max feature normalization. Finally, we deployed the backend using FastAPI. This prototype provides a secure, transparent baseline for future clinical testing. We also added Explainable AI (SHAP/LIME) so users understand their recommendations, and we map out how to link the system to Continuous Glucose Monitoring (CGM) devices moving forward.},
        keywords = {Machine Learning, K-Nearest Neighbors, Neural Collaborative Filtering, Large Language Models, Explainable AI, Personalized Nutrition.},
        month = {April},
        }