Abstract:
Blood donation is an activity of distributing blood or blood-based products from
one person to another person's circulatory system. As a result of decreased blood
supply, a large supply of blood is needed to anticipate emergency situations. The
purpose of this research is to simplify the process of examining prospective blood
donors, so that the donation process can be more effective and efficient. This
research uses the k-Nearest Neighbor (k-NN) method, using data from UDD PMI
Padang City in 2021 as many as 623 data. The research was conducted by looking
for the best model from the data which is divided into nine partitions, each of the
nine partitions is calculated with neighbor values k = 3 to k = 15. the best model is
a model with a data partition of 80%:20% with 498 training data and 125 data
testing data using k=9 which produces an accuracy value of 92%.