Penerapan Artificial Neural Network dan Rule Based Classifier untuk Mengklasifikasikan Pendonor Darah Potensial pada Sistem Broadcast Pendonor

Widya Arditanti (Program Studi Teknik Informatika, Universitas Kristen Petra Surabaya)
Andreas Handojo (Program Studi Teknik Informatika, Universitas Kristen Petra Surabaya)
Tanti Octavia (Program Studi Industri, Universitas Kristen Petra Surabaya)

Jurnal Infra
Vol 10, No 2 (2022)

One of UTD PMI Surabaya’s task is to provide safe and quality blood when blood is needed in an emergency. The availability of blood at UTD PMI Surabaya can be erratic, because it depends on the number of donors that fluctuates and the storage time of blood is not long. Therefore, UTD PMI Surabaya needs a system to invite potential donors to meet blood needs when needed in an emergency, by minimizing blood wasted. The classification model and the creation of a recommendation system will produce a list containing donors who have been sorted by priority. Testing was carried out by dividing the data according to the conditions of the data collection environment (before the pandemic, during the pandemic and a combination of before and during the COVID-19 pandemic). The highest MRR value was obtained from the ANN model made from a combined data of 90% classification results using RBC and fake data. The accuracy value obtained from the model is 91.13% for training and 91.83% for testing. The resulting MRR value is 8.07 x 10-4 .