Using Machine Learning Approaches for Prediction of the Types of Asthmatic Allergy across the Turkey

Abstract

Nowadays, allergy is thought to be an important cause of frequent occurrence of diseases in the society we live in. Hence, finding out relation between patient characteristic variables such as age, sex and type of allergic diseases such as asthma, allergic rhinitis, food allergy, allergic dermatitis and so on is the main objective among allergy researchers. In this study, we propose to design an intelligent diagnostic assistant for prediction of the type of an allergic disease across Turkey automatically by using well-known machine learning algorithms such as Decision Tree, Logistic Regression, Support Vector Machines (SVM), K Nearest Neighbor (kNN) and ensemble classifiers. In experiments, an allergic diseases dataset, which is taken from Kocaeli University Research and Application Hospital, is utilized. As a result, in detecting 18 different allergy diagnoses, the maximum accuracy rate of 77% is achieved with majority voting.