Early Prediction of Childhood Obesity Using Machine Learning Techniques

The aim of this research work is the early prediction of childhood obesity after the age of three years from available clinical records of patients. Nowadays, child obesity is a highlighted research area as excessive body fat harmfully affects a child’s health. Obese children have more risk of suffering from health problems such as heart diseases, type 2 diabetes, cancer, and osteoarthritis in their adulthood. Thus, early prediction of childhood obesity is essential for fat and overweight babies. In this paper, we have proposed a prediction model for this purpose. Analyses of three different machine learning methods: SVM, KNN, and ANN for establishing accuracy in the prediction model have been done. From the result analysis, it can be established that a prediction model based on machine learning techniques can be used to predict obesity in children after the age of two years.