Application Tool based on C4.5 Decision Tree for Diagnosing Diabetes Infection Symptoms

Abstract

Abstract—Diabetes is a fatal disease which can lead to many other dangerous illnesses such as blindness, hypertension, kidney failure, heart attacks, and gangrene.   Nowadays, the intelligent diabetic systems exploit important tools to both medical industry and diabetes patients due to their crucial role in improving the quality of healthcare in many ways. Such systems can be considered as a very helpful tool to allow the physicians and the doctors intervene to find the proper treatment for the patient with all kinds of conditions. This paper therefore presents an application of intelligent system for diabetes detection symptoms to support and give advice to clinical management and patients. It basically relies on how to detect and find the Probability of Infection Diabetes. Hence, if a person suffers from the symptoms of a group, a patient will be referred to the possibility of diabetes. System diagnostics are examined based on the algorithm of Szajnar and Setla. It starts whether there is infect or not infect/doubt in the possibility of injury. When the person is doubted with the probability of injury, the probability of injury with symptoms can pass through a set of resulting rules from the C4.5 decision tree algorithm. The results reveal the finding ratios of Incidence percentage by the sum of the values of the outputs of the rules derived. It was ascertained the validity of the results by comparing them with the Indian diabetes database whereas if the injury rate of less than 50 is not infected and greater or equal to 50 is infected.  Consequently, the implementation of this expert application tool shows very good results.