Machine Learning Approaches for Prediction of Nutrition Deficiency among Women of Different Age Groups

Malnutrition has been a major health concern widely especially when it comes to Asian women. Due to the lack of access to healthcare facilities for the women of underdeveloped and developing countries, the women of these regions suffer the most. For this purpose, an effort is made to predict the nutritional status of the women in one such country Pakistan. The research aims to build an intelligent model for the prediction of nutritional deficiency among women. Machine learning algorithms namely Support vector machine, random forest, Logistic Regression, KNN, and Naïve Bayes have been used to build the nutritional predictor in relevance to Pakistani women. The result shows that the Random Forest Model has the highest accuracy and is the most suitable one for future predictions. The model would enable us to know the health condition in particular of Pakistani women. The testing of the model on a similar dataset has further confirmed the reliability factor.