Optimasi Parameter Support Vector Machine (SVM) dengan Particle Swarm Optimization (PSO) Untuk Klasifikasi Pendonor Darah Dengan Dataset RFMTC

I Gusti Ngurah Ersania Susena (Fakultas Ilmu Komputer, Universitas Brawijaya)
Muhammad Tanzil Furqon (Fakultas Ilmu Komputer, Universitas Brawijaya)
Randy Cahya Wihandika (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 12 (2018): Desember 2018

Blood donation is one of voluntary humanitarian activities. Blood is one of the most important substances that humans have in the human life cycle. In carrying out blood donation activities, monitoring the stock availability of blood bags is usually a major problem. To know ammount stock of blood bag we need a system that can predict the behavior of blood donors. RFMTC (Recency, Frequency, Monetary, Time, Churn Probability) is a modified RFM method in order to see the behavior of donors who can donate their blood or not to donate again. Therefore, SVM-PSO method needed to know classification of blood donors behavior. With SVM techniquesto find hyperplane that is the dividing line between data classes. Then the PSO technique to find the range of input parameters that SVM needed to get the optimal hyperplane value. This research uses 748 data from UCI dataset with 4 main features and 2 classes. Based on the test that has been done obtained the accuracy of 90% with the value of learning rate SVM small and the value of the number of PSO particles are low.