Interactive Machine Learning Framework for Predicting Asthma Health Conditions Using XGBoost

Abstract

Asthma is a far reaching non-communicable sickness that affects individuals of any age and is a significant danger to human wellbeing worldwide, especially in low and middle-income countries. Its prevalence is substantial, and it remains one of the major contributors to global death rate. A substantial number of asthma development prediction models, and machine learning based approaches, exist. The problem associated with conventional methods of asthma prediction is low predictive accuracy of the model. The objective of this study is to develop an interactive Machine Learning (iML) framework for predicting Asthma Health conditions in selected African countries. The model was tested using dataset obtained from National Survey of Children's Health (NSCH). The machine learning technique used was XGBoost, and model development including preprocessing, training, and tested were carried out using Python programming language on Google Collaboratory (Collab) environment. The results of benchmarking of the performance accuracy of XGBoost shows 90.6% which is higher than the existing results in literature for Decision Tree (90.49%), KNN (90.28%), Naïve Bayes (82.7%). The study found out that the model developed by the researchers performed comparatively well in predicting asthma when compared with some existing state-of-the-art machine learning algorithms.