Open AccessOptimization of C4.5 algorithm using meta learning in diagnosing of chronic kidney diseases
Author(s) -
Aldi Nurzahputra,
Much Aziz Muslim,
Budi Prasetiyo
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1321/3/032022
Subject(s) - confusion matrix , kidney disease , decision tree , computer science , data mining , big data , disease , confusion , algorithm , machine learning , artificial intelligence , medicine , pathology , psychology , psychoanalysis
The accuracy in diagnosing a disease is very important and crucial for the hospital institution. The big patient data can be processed into information to diagnose a disease. One of diseases can be diagnosed is chronic kidney disease. The existing data of chronic kidney disease patients can be used for data processing. Data processing is included in the process of data mining. One method that can be implemented to disease diagnosis, namely classification. The algorithm used in this research is decision tree namely C4.5 algorithm. It can be applied in classifying patient data. This study used Chronic Kidney Disease (CKD) dataset. The purpose of this study is optimizing the accuracy of C4.5 algorithm by applying meta learning in diagnosing CKD by comparing the results before and after MultiboostAB and Bagging applied. The validation used 10 fold cross validation. The accuracy is measured by confusion matrix. The combination of C4.5 and MulitboostAB obtained 99.5% accuracy which increased 0.5% from accuracy of C4.5 standalone. Then, C4.5 and Bagging obtained 100% accuracy. The result of this research is the application of bagging on C4.5 algorithm is good in optimizing accuracy.