<![CDATA[Medical record data are special patient records, often medical record data only becomes data that accumulates and is not searched to produce useful knowledge for hospitals. This study aims to determine the disease classification model based on piles of medical record data, using one of the methods in data mining. To achieve the research objectives, 4 attributes were selected according to the medical record data. The medical record data in question consists of disease diagnosis attributes based on the International Classification of Diseases-10 (ICD-10), gender, age of the patient and month of admission to the hospital. The method used in this study is the C4.5 algorithm method using the international disease code attribute as the destination label attribute for 21 international disease groups, namely: A00-B99 to Z00-Z99. In this research, the C4.5 algorithm can represent 7 attribute values for the disease code, namely A00-B99, C00-D89, I00-I99, O00-O99, P00-P96, S00-T98 and Z00-Z99. The conclusion of this study is that the C4.5 algorithm is less than optimal in producing classification of medical record data because the number of destination classes or class labels is very large and the percentage of data read is less than 50%. The resulting disease classification is only 7 classes out of 21 overall classes according to the international disease code.]]>
Copyrights © 2022
