<![CDATA[Blood donation is a process of taking blood from a person voluntarily to blood transfusions for patients in need. Blood from donors can't be used after 42 days. The only way to meet the demand for blood bags are having regular blood donations from a healthy donor. In Indonesia 2013, there is 2,476,389cc shortage of blood bags, where the ideal blood availability is 2.5% of the population. These problems required a system that can predict the behavior of donors in order to anticipate the shortage of blood bags. Regency, Frequency, Monetary, Time, Churn Probability (RFMTC) is a modification of the RFM method modified that used to predict the blood donor behavior to donate or not to donate bloods. The method for classifying the behavior of donors in this research are Support Vector Machine (SVM) method. Data that was used in this research is 748 which is divided into training data and test data. The accuracy result got best accuracy based on 50%: 50% data ratio, using linear kernel and parameter value λ (lambda) = 2, Gamma (γ), = 0.5, Epsilon (ε) = 0.005, and C (complexity) = 20. The result of SVM method accuracy on blood donor classification is 72.64%.]]>
Copyrights © 2018
