An Improved C4.5 Data Mining Driven Algorithm for the Diagnosis of Coronary Artery Disease

• Main Authors: Haruna, A.A., Muhammad, L.J., Yahaya, B.Z., Garba, E.J., Oye, N.D., Jung, L.T.
• Format: Conference or Workshop Item
• Published: Institute of Electrical and Electronics Engineers Inc. 2019
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087076704&doi=10.1109%2fICD47981.2019.9105844&partnerID=40&md5=ca28c484c55fb7565b5d384818a8e2b5; http://eprints.utp.edu.my/23669/

Coronary artery disease (CAD) is one of the deadly diseases in the world, especially in developed countries. This disease is not epidemic but it re-mains the single most common cause of death. This research used an im-proved C4.5 data mining algorithm for the diagnosis of CAD. A performance evaluation of the improved algorithm was carried out against the traditional C4.5 Algorithm. Consequently, the improved C4.5 data mining algorithm has shown better performance with an overall accuracy of 97.23 , 97.03 specificity, and 96.39 of sensitivity. The improved algorithm built a tree with twenty-seven leaves and forty-seven sizes, which can be converted into the production rules for knowledge base of expert system to diagnose CAD. This helps in addressing the problematic bottleneck of knowledge acquisition process in expert system for diagnosis of CAD. © 2019 IEEE.