<![CDATA[According to WHO, 17 million people die every year due to diseases of the heart and blood vessels around the world. Coronary Heart Disease (CHD) alone holds the title as the cause of death and disability in the world number one. It will be applied algorithm C4.5 decision tree feature selectionbased Particle Swarm Optimization (PSO) to find out how accurate the PSO feature selection to improve the accuracy of the C4.5 decision tree in predicting heart disease.The results for the accuracy of classification algorithm C4.5 worth 81,25%, whereas the accuracy for C4.5 classification algorithm based on PSO is worth 93,75% the value that is equal to 12,5% accuracy. While evaluation using ROC curve for both, the value of AUC by ROC curve for C4.5 classification algorithm is worth 0,718 with Fair diagnosis classification level, wheras for C4.5 classification algorithm based on PSO is worth 0,855 with Good Classification diagnosis rate, the difference in AUC values is 0,137. It can be concluded that the application of particle swarm optimization techniques can improve the accuracy of the algorithm C4.5.]]>
Copyrights © 2014
