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{195695,
        author = {Hrushikesh Akkinapelli and Ankur Manoj Reddy and Balyala Ajay and Bojja Chandu},
        title = {Health-Aware Food Recommendation System Integrated with Food Delivery Applications},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14610-14615},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195695},
        abstract = {Online food delivery platforms have grown explosively over the past decade, reshaping how people make daily dietary choices. Yet the recommendation engines powering these platforms still treat every user as a generic consumer, optimising for clicks and ratings while ignoring the medical reality that roughly one in three adults worldwide lives with a chronic condition that makes certain food items genuinely dangerous. This paper describes a Health-Aware Food Recommendation System that corrects this oversight. The framework layers four complementary components on top of a standard food delivery pipeline: a matrix factorisation model that learns individual taste preferences from interaction history; a nutrition-based health scoring module that assigns every food item a continuous risk value between 0 and 100; a rule-driven constraint filter that removes items breaching disease-specific dietary thresholds before any recommendation is generated; and a Q-learning reinforcement agent that refines the ranking policy after every user feedback event. Experimental evaluation on a large-scale food interaction dataset showed that the hybrid system raised Precision@10 from 0.71 to 0.86, cut the fraction of medically unsafe suggestions from 18.4 % to 2.1 %, and achieved a cumulative RL reward of +0.73 over the evaluation period. The system is deployed as a full-stack web application using React on the frontend and Flask on the backend, with real-time food API integration and cloud-based data persistence.},
        keywords = {collaborative filtering, dietary constraint filtering, food delivery integration, health-aware recommendation, nutritional health scoring, personalised nutrition, Q-learning adaptation, reinforcement learning.},
        month = {April},
        }