Open AccessApplication of blood donor routine detector using K-Nearest neighbors
Author(s) -
Yanuar Nurdiansyah,
Priza Pandunata,
Nova Prasetyo,
Agus Trihartono,
Flora Grace Putrianti,
Feri Wijayanto
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/293/1/012042
Subject(s) - computer science , k nearest neighbors algorithm , simple (philosophy) , training set , sample (material) , waterfall model , set (abstract data type) , class (philosophy) , data set , data mining , medicine , artificial intelligence , software , operating system , philosophy , chemistry , epistemology , chromatography , programming language
The application of blood donor routine detection using the K-nearest Neighbors method at UTD PMI (Unit Transfusi Darah, Palang Merah Indonesia , Indonesian Red Cross, Hereafter PMI) is a system that aims to provide the class status of the donors aimed at facilitating the employee in selecting the right donor to be recontacted. The K-Nearest Neighbors method is used to calculate the distance of the test data with the training data. The K-nearest Neighbors method is chosen because it has a simple algorithm, works based on the shortest distance from the test data sample to the training data sample to determine the distance. The creation of this system is built by adapting the waterfall model. The objective is that the system is able to implement the K-nearest Neighbors method to help determine the class status of the donors and that the system can send SMS gateway to the donors that donor information can be publicly published. The test calculation uses full train full set testing on all training data and the result of K5=84 %, K7=88 %, K9=87 %, K11=82 %, K13=77 %, K15=75 %. The highest result is on the K7.