MEDICAL INFORMATICS IV
COMPARATIVE ANALYSIS OF MACHINE LEARNING METHODS IN DISEASE DETECTION APPLICATION TO DIABETES MELLITUS AND HEART DISEASES
Yayıncı:
İstanbul Üniversitesi Yayınları
The rapid advancement of technology and developments in computer science have brought about significant changes in the field of health, as in many other fields. Artificial intelligence and machine learning have become an important trend in early recognition of the effects that cause diseases, investigating the symptoms of the disease, making the correct diagnosis, and classifying the disease, improving health processes, early diagnosis of diseases, preventing the spread of diseases, reducing health costs, and increasing the effectiveness and quality of health services. has arrived. In this study, machine learning methods were used to detect diabetes and heart diseases, which are among the world’s deadliest diseases and threaten public health. The performances of machine learning algorithms in correctly diagnosing diseases were compared. In this study, by applying the Smote technique, better accuracy results were obtained with Gradient Boosting (GB), Random Forest Classifier (RF), Extreme Gradient Boosting (XGB) methods
