A research submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Information Management, School of Engineering and Technology

Research studies project report (M. Eng.) - Asian Institute of Technology, 2018

In the recent era, expert systems have been developed in large scale which have affected all sectors of human life especially in medicine. The purpose of this study is to analyze risk of diabetes in patients and predict its seriousness by using the classification techniques depending on attained medical datasets. Basically, diabetes is a chronic disease which severely affects the body's ability to use blood sugar properly. Moreover, its common symptoms may include increased thirst and urination, blurred vision, and fatigue. The scary facts about diabetes is its rapid prevalence worldwide as well as its unawareness among the people. Hence, medical practitioners suggest if diabetes could be diagnosed early, it may save lots of lives across the world. For this reason, an expert system is developed that can diagnose diabetes by considering the dataset{u2019}s basic attributes such as age, plasma glucose concentration, triceps skin fold thickness, 2-hour serum insulin, body mass index, and diastolic blood pressure. Notably, this system can predict risk of diabetes by applying decision tree and rule induction data mining techniques to achieve results. Comparatively, decision tree algorithm proved to be better in performance and accuracy than rule induction algorithm. PHP, HTML, phpMyAdmin, Tomcat, Atom IDE were used for the development and implementational purposes. Lastly, system was evaluated with both testing dataset{u2019}s instances and users.