HYBRID METODE BOOSTRAP DAN TEKNIK IMPUTASI PADA METODE C4-5 UNTUK PREDIKSI PENYAKIT GINJAL KRONIS

Ahmad Ilham(1*)


(1) 
(*) Corresponding Author

Abstract


Missing values is a serious problem that most often found in real data today. The C4.5 method is a popular classification predictive modeling used because of its ease of implementation. However, C4.5 is still weak when testing data that contains large missing. In this study we used a hybrid approach the bootstrap method and k-NN imputation to overcome missing values. The proposed method tested using Chronic Kidney Disease (CKD) data, and evaluated using accuracy and AUC. The results showed that the proposed method was superior in overcoming missing values in CKD. It can be concluded that the proposed method is able to overcome missing values for chronic kidney disease prediction.

Keywords


Missing Values, Bootstrap; K-NN; Chronic Kidney Disease Prediction

Full Text:

PDF

Article Metrics

Abstract view : 250 times
PDF - 101 times

DOI: https://doi.org/10.26714/jsunimus.8.1.2020.%25p

Refbacks

  • There are currently no refbacks.


Copyright (c) 2020 Jurnal Statistika Universitas Muhammadiyah Semarang

Editorial Office:
Department of Statistics
Faculty Of Mathematics And Natural Sciences 
Universitas Muhammadiyah Semarang

Jl. Kedungmundu No. 18 Semarang Indonesia

Published by: 
Department of Statistics Universitas Muhammadiyah Semarang

View My Stats

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.