A Machine Learning Model for Malnutrition Detection in Preschool Children

• DOI: 10.52783/jisem.v10i41s.7851
• Publication Date: 2025-05-03T09:49:34Z
• Authors: Amol Avinash Shinde

The use of artificial intelligence techniques has introduced a new dimension to the health-care sector, particularly in diagnosing diseases their early stages, allowing patients and workers take corrective action. study discusses machine learning algorithm created detect classify malnutrition preschool children. Malnutrition affects significant number children various nations, many whom are preschoolers. Early identification child will allow parents monitor child's health. model also proposes dietary plans based on type diagnosed. Introduction -Between ages one five, start develop diet habits, often influenced by exposure tastes market foods. It can lead health issues like obesity or insufficient nutritional intake. arise from food that does not meet child’s needs an imbalanced diet. treatment crucial avoid long-term problems affecting overall development. A well-balanced is vital prevent chronic conditions. Parents try provide adequate nutrition, but deficiencies development stages cause stunted growth, weakened immunity, other issues. Objectives- primary aim early. By predicting malnutrition, assists taking preventive actions managing effectively. Additionally, personalized diagnosed malnutrition. emphasizes providing simple, questionnaire-based system for remote rural parents, making healthcare advice easily accessible. Methods –The was Exploratory Data Analysis (EDA), attribute selection, training, evaluation. Initially, twenty-eight attributes were shortlisted clinical research, then reduced twenty-one essential attributes. Various algorithms tested using WEKA tool, Logistic Model Tree (LMT) selected its high accuracy. trained with 70% dataset 30%. two-phase implementation done first (undernutrition micronutrient deficiency). Results achieved accuracy 91% detecting two-step diagnosis improved detection undernutrition deficiencies. Grouping related medical factors further increased 95% second phase. successfully classified as healthy, undernourished, over-nourished, suffering deficiencies, suggested appropriate diagnosis​. Conclusions application models supports intervention better outcomes. developed accurately predicts preschoolers provides user-friendly platform especially those areas. suggesting customized plans, helps manage effectively, reducing risk problems.