A Machine Learning Framework for Evaluating Nutritional Status Disparities Based on Income Among Adolescents in Bangladeshi Sub-Urban Areas

The present study, using a machine learning model, ascertains the relation between nutritional deficiencies and income disparities among adolescents in sub-urban Bangladesh. In total, data from 400 adolescents was analyzed, having income as the independent variable and nutritional score as the dependent variable. Linear Regression, the result of Mean Squared Error (MSE) - MSE: 0.146, R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>: 0.0001, Random Forest - MSE: 0.1529, R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>: -0.0447, and XGBoost - MSE: 0.1483, R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>: -0.0131. These results were obtained for income versus nutritional status: weak correlation with Pearson 0.1676 and p-value 2.54e-5. This may indicate that the other factors are more important. Further investigation on other factors might be required to understand adolescent nutrition.