Integrated Data Mining And Knowledge Based System For Prediction And Treatment Advice Of Diabetes
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Diabetes Mellitus Is A Group Of Metabolic Diseases Characterized By Hyperglycemia Resulting From Defects In Insulin Secretion, Insulin Action Or Both. It Is Divided Into Two Major Type?�?S Type1 And Type2 Diabetes. During The Past Decade, Diabetes Mellitus Has Emerged As An Important Clinical And Public Health Problem Throughout The World, Especially In Developing Country Like Ethiopia (800,000 Patients In 2000 E.C) .Extracting Implicit Knowledge From Domain Expert About Diabetes Was So Difficult In Order To Get Knowledge About Diabetes.The Main Objective Of This Study Is To Develop A Predictive Model And Integrate The Model With Knowledge-Based Diabetes Prediction System That Provides Advisement And Decision Support For The Users.To Build The Predictive Model Dataset Is Pre-Processed For Missing Values, Outliers, Noise And Errors, Based On The 12,139 Sample Data Gathered From Black Lion Diabetes Clinic Center. In This Study The Model Is Experimented Using J48 Decision Tree, Jrip, PART, LMT And Reptree. For The Knowledge-Based Part, Knowledge Is Acquired From Experts And Documents, The Knowledge-Based System Integrating With Data Mining Was Developed Using WEKA, Java, Swi-Prolog Tools.In This Study As Compared To Other Algorithm, The Performance Of J48 Decision Tree Reveals That 93.84% Correct Results Are Achieved For Developing Classification Rules That Can Be Used For Prediction And The System Achieved 91.6% Of The User Acceptance. Thus, Knowledge Discovered With This Algorithm Is Used To Integrate With Knowledge-Based System Using Java. As A Result, The Knowledge-Based System Is Used To Predict Based On The Attribute Values.The Researcher Concludes That Diabetes Prediction And Advisement Should Be Implemented At Any Time To Be Effective It Should Be Supported By Data Mining Technology To Overcome On Problem Of Diabetes. However, Further Investigation Should Be Required For The Future To Significantly Support This Struggle.
