Diagnosis of respiratory diseases for children using machine learning

respiratory diseases are among the widespread diseases that affect child in abundance around the world, and which can increase the number of child deaths due to the speed of their spread and thus exhaust the global health institutions. Respiratory diseases vary according to their symptoms, and some of them may share some of the symptoms, the most famous of them are asthma, bronchiolitis, and pneumonia. Diagnosing these types of diseases and differentiating them based on clinical symptoms is not easy, especially for junior doctors, which may be inaccurate diagnosis and thus endanger the lives of children. The main objective of this study is to improve the diagnosis These three diseases are for children under two years of age and differentiate between them using machine learning models to classify the real data set based on the attributes provided by the pediatric consultant to help the junior doctors in distinguishing between these diseases. Two machine learning models were applied to the real data set based on the evaluation metrics. The results showed the superiority of the Naïve Bayes classifier, with an accuracy of 99.7093, over the second classifier Logistic Regression and therefore, it was adopting the first classifier as a classification model for our study.