STUDI KOMPARASI METODE NAIVE BAYES DAN K-NEAREST NEIGHBOR (KNN) UNTUK MEMPREDIKSI PENDONOR DARAH POTENSIAL (STUDI KASUS: BLOOD TRANFUSION SERVICE CENTER)

Nur Faizah, NIM.: 16650012 (2020) STUDI KOMPARASI METODE NAIVE BAYES DAN K-NEAREST NEIGHBOR (KNN) UNTUK MEMPREDIKSI PENDONOR DARAH POTENSIAL (STUDI KASUS: BLOOD TRANFUSION SERVICE CENTER). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (STUDI KOMPARASI METODE NAIVE BAYES DAN K-NEAREST NEIGHBOR (KNN) UNTUK MEMPREDIKSI PENDONOR DARAH POTENSIAL (STUDI KASUS: BLOOD TRANFUSION SERVICE CENTER)) 16650012_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (STUDI KOMPARASI METODE NAIVE BAYES DAN K-NEAREST NEIGHBOR (KNN) UNTUK MEMPREDIKSI PENDONOR DARAH POTENSIAL (STUDI KASUS: BLOOD TRANFUSION SERVICE CENTER)) 16650012_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

The lack of supply of blood bags in Indonesia every year results in a high mortality rate in pregnant and childbirth mothers. The importance of fulfilling blood bag stocks can improve health care and save a person's life. Donated blood cannot be used after 42 days, so to meet the minimum stock of blood bags, then potential blood donors should be known. Predictions can be made using methods in data mining that have high accuracy. Predictions are made using blood transfusion service center data from UCI Machine Learning. The data amounted to 748 with 4 donor condition attributes to determine predictions that were then evaluated correlations between dependent attributes. Predictions were made using Naive Bayes and K-Nearest Neighbor which then compared the accuracy values of each method to get the highest score. Based on the test using the Naïve Bayes algorithm, the algorithm produces 76% accuracy for the RMT dataset using preprocessing Z-Score Normalization and Minmax Scaler while the K-Nearest Neighbor (KNN) algorithm produces an accuracy of 79.3% on a comparison of 20% test data and 80% training data using dataset Recency, Monetary and Time, and Z-Score Normalization for data preprocessing. So it can be concluded that the K-Nearest Neighbor (KNN) algorithm is the most relevant algorithm used to predict potential blood donors.

Share this knowledge with your friends :

Actions (login required)

View Item