Research Article
Predicting Healthy Diets: A Machine Learning Approach for Personalized Nutrition
@INPROCEEDINGS{10.4108/eai.28-4-2025.2358050, author={Kallakuri Narendra Sharma and Guthurthi Lalitha and Gorla Darshini Sai and Lalam Ramya and Uppada Siddhartha Reddy and Bagadi Lasya Priya}, title={Predicting Healthy Diets: A Machine Learning Approach for Personalized Nutrition}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part II}, publisher={EAI}, proceedings_a={ICITSM PART II}, year={2025}, month={10}, keywords={personalized diet recommendation machine learning data preprocessing feature engineering ensemble learning xgboost lightgbm random forest knn stacking}, doi={10.4108/eai.28-4-2025.2358050} }- Kallakuri Narendra Sharma
Guthurthi Lalitha
Gorla Darshini Sai
Lalam Ramya
Uppada Siddhartha Reddy
Bagadi Lasya Priya
Year: 2025
Predicting Healthy Diets: A Machine Learning Approach for Personalized Nutrition
ICITSM PART II
EAI
DOI: 10.4108/eai.28-4-2025.2358050
Abstract
Personalized diet recommendation systems are increasingly vital in driving better food choices. This work presents a comprehensive architecture for predicting diets and thereby recommending effective nutrition schedules based on the predictions. The proposed structures employ an extensive dataset preparation operation to go through the diet dataset generation, cleansing, normalization, feature engineering, and outlier treatment for better quality inputs. Secondly, the versatile ensemble framework combines XGBoost, LightGBM, Random Forest, and KNN models to provide the requisite impetus. Thirdly, the predictions are combined in a multi-tier stacked ensembling fashion to obtain better precision and credibility. The results show the stacking ensemble methodology achieves approximately 96% accuracy, and the prototype can suggest personalized and decisive recommendations in response to dietary habits.
