Extreme Programming Implementation for Diabetes Mellitus Early Detection Expert System Using Forward Chaining Method
Nurhayati, Akhiat, Ashar Abilowo, Fransiskus Panca Juniawan, Dwi Yuny Sylfania

Poltekkes Kemenkes Pangkalpinang
Universitas Bangka Beleitung

Abstract

Diabetes mellitus is a chronic disease that affects millions of people worldwide and is a major cause of morbidity and mortality. Early detection and intervention are crucial in managing the disease and preventing complications. In this paper, we propose an expert system for the early detection of diabetes mellitus using the forward chaining method, which developed with extreme programming. By combining the knowledge of team research and doctor interna as internal medicine specialist, we agreed that this system will have 17 symptoms. This system will utilize data from patient symptoms to generate a diagnosis quickly and accurately by using the Forward Chaining approach, which has ability in provide a lot of information from small data. In addition, we use extreme programming method in order to develop the system that can accurately and efficiently diagnose diabetes in its early stages. For the functional test, we conducted Blackbox Testing, with the result that all features were 100% working exactly as expected. Moreover, the next testing were 88,89% for Confusional Matrix Testing and 90 % for Human Expert Validation Testing.

Keywords: Diabetes Mellitus, Early Detection, Expert System, Extreme Programming, Forward Chaining

Topic: Electrical, Electronic, and Computer Engineering