A Comparative Study of Diabetes Classification Based on Machine Learning

Diabetes is commonly acknowledged as one of the most dangerous chronic illnesses. Diabetes can potentially cause a range of severe complications, such as blindness, kidney disease, heart attacks, and other related illnesses. Better technology is needed to achieve better patient monitoring, drug management, monitoring, and early medical intervention. In recent years, there has been a lot of emphasis on diabetes prediction via classification methods. The early diagnosis of the disease with its symptoms could improve treatment and reduce the risks of developing any dangerous consequences. This study applies several machine algorithms techniques to a dataset of 16698 instances, which covers the essential body features of multi-aged people. Three different machine learning algorithms, namely k-nearest neighbor (KNN), Decision Tree (DT), and Naive Bayes (NB), were employed. The main focus is to enhance a system that could approach what real-life situations usually demand from accuracy, sensitivity, and complexity time. The experimental results showed that KNN and DT maintained a 100% sensitivity with 99.64% and 99.61% accuracy values, respectively. In addition, the three classification systems could predict diabetic records with good time complexity.