Feature analysis of food products and health benefits classification using sequence ranking with the gradient boosting method

Healthy foods must be marketed to kids to influence their nutrient choices. Check the food source for newborn nutrition and health advantages. Ensemble learning solves issues and improves predictions. Machine learning targets obesity and metabolic health nutrition prediction. Previous methods were inaccurate, showing inadequate product ingestion and long health benefit forecast timeframes in digital promotion. A model based on childrens nutrition information (protein, sugar, fat, iron, folic acid, calcium, and immunity) predicts the nutrient composition of prepared foods using SRCFA-NBGB. Starting with kid food nutrition, we used standard repository data. Data pre-processing begins with Min-Max Z-score normalisation to remove null values. Step two uses sequential ranking clustering analysis to choose features with maximum weights. Use a subset of attributes to determine dietary nutritional threshold margins and maximum feature values (protein, sugar, fat, iron, folic acid, calcium, and immunity). Phase three, feature weights validation, uses stratified K-fold validation to discover the maximum threshold values by dividing the data into k nearly equal-sized groupings. Final phase, NBGB classification, predicts nutritional quality and recommends healthy food for kids. NBGB forecasts use the recommended foods accuracy, precision, recall, F1-score, error rate, and temporal complexity. The confusion matrix metrics above are superior.