Claim Missing Document
Check
Articles

Found 1 Documents
Search

Penerapan Metode Decision Tree dan Algoritme Genetika Untuk Klasifikasi Risiko Hipertensi Selly Kurnia Sari; Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 3 (2019): Maret 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (314.55 KB)

Abstract

Hypertension is one of the third most deadly diseases after stroke and tuberculosis which reached 6.7% of the total age in Indonesia. This shows that hypertension in Indonesia needs serious attention as an effort to deal with this problem. Handling is needed as a step in early detection of hypertension. In relation to the classification for detection of hypertension, one method that can be used is the Decision Tree (DT). But in the previous study DT method produced a fairly low accuracy. To optimize the accuracy of the DT method, Genetic Algorithm (GA) is used. GA is used to generate new rules. To find out the difference in the results of accuracy, a comparison test of DT-C4.5 and DT-GA is conducted. The test uses the same test data. The test results show the DT-GA algorithm produces the highest average accuracy of 84%. While the DT-C4.5 algorithm produces the highest average accuracy of 70,5%. The best parameters to produce the best accuracy are population size 60, Cr = 0,3, Mr = 0,7, with the maximum number of generations used is 10. From the results of these tests it can be concluded that GA can be used to generate new rules from DT