Recommendation Method Based on Nutritional Density for Intake Order of Foods Detected by Traditional Machine Learning

Abstract

Most of the current food recommendation methods Forget about calorie or glycemic index, most people tend to focus primarily. looking the overall nutritional quality of foods. To address With this limitation, this study suggests a Nutritional Quality Score.This limitation, the present study suggests a Nutritional Quality Score. The (NQS) framework, which determines foods according to the balance, is used to assess them. The balance between beneficial nutrients (protein, fibre) and less desirable nutrients (fat, sugar). Sugars, sodium and calories (nutrients). Assume that a person is taking in the Daily Food & The Nutrition Dataset contains 10,000 dietary records and 2,039. distinct food items, nutritional attributes were preprocessed and Were used to define NQS values for food ranking. Traditional machine learning algorithms, including Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN), were two of the most popular classifiers. used to determine consistency and predictive power of The proposed NQS framework. Experimental results show that The highest accuracy was found in the model of the Random Forest. an R2 score of 0.9785 and a Spearman correlation of 0.9955. The plant-based method suggested in the framework allows nutrient-density based foods to be produced. Ranking and intake-order recommendation while preserving The interpretability and simplicity of computation. The findings Show what’s possible with nutrition-density driven. interventions to promote healthier food choices and diets.strategies to encourage healthier food choices and diets. Future work will focus on personalization, user-centered evaluation, and validation using real-world dietary outcomes.