Main Article Content
Abstract
Potential blood donors are blood donors who can donate their blood back after success through 2 stages of blood donation such as the physical health test (active) and the screening test (laboratory test). The purpose of this study are to obtain an application that can be used to predict potential blood donors who will donate their blood back at the PMI Palu, Sigi and Donggala Blood Transfusion Units, and to obtain their level of accuracy using the Learning Vector Quantitation algorithm. This prediction application for potential blood donors makes it easier for the public to know whether they can donate their blood or not. Classification is done using 300 data consisting of 70% training data and 30% testing data. The data used in this study are data taken in 2018. The accuracy of the best weighting in stage 1 is 95.56% obtained using the training rate (α) of 0.1≤α≤0.25 and the rate reduction training (decα) which is varied. While the best weighting results in stage 2 have an average accuracy rate of 100% obtained by using a training rate (α) of 0.000001≤α≤0.5 and a reduction in the rate of training (decα) which varies.