KLASIFIKASI DATA REKAM MEDIS BERDASARKAN KODE PENYAKIT INTERNASIONAL MENGGUNAKAN ALGORITMA C4.5

Abstract

This study aims to determine the classification of disease models based on the data stack of medical records, using one of the methods in the data mining algorithm C4.5. To achieve the research goal is then selected four attributes appropriate medical records consisting of attributes Diagnosis of disease based on the International Classification of Diseases-10th (ICD-10), Gender, patient age, Month patient admission to the hospital. The results show that there are 5 types of disease classification are A00-B99, I00-I99, J00-J99, O00-O99 and Z00-Z99. A00-B99 disease generally affects men with young and adult age categories, women with older age category, the category of infants and children only occurred in January, March, April May. I0-I99 disease generally affects men with older age category. J00-J99 diseases commonly suffered by men with age categories of infants and children in November. O00-O99 illnesses commonly suffered by women with young and adult age categories. Z00-Z99 disease usually affects infants and children in February, June, July, August, September, October, December, whereas in November suffered by infants and children with the female gender. C4.5 algorithm generates a classification less than the maximum in the medical records for the number of classes or class label purposes very much and the percentage of data that is read is less than 50%. Classification of diseases produced only 5 classes of 21 overall class by international disease code.