Abstract
High-risk of pregnancy refers to a situation where pregnancy will have a negative impact on the safety of the mother and baby. Since the beginning of pregnancy, high-risk pregnancy can be predicted by various factors such as the physical and psychological condition of the pregnant woman, nutritional intake, and congenital diseases. According to WHO, Indonesia ranks 5th in premature birth rates with 675,700 babies and this figure is 15.5% of the total birth rate in Indonesia. Estimates of high-risk pregnancies can be observed from patient medical record data, in this case, pregnancy data from pregnant women. Data that is processed into knowledge can be processed through the data mining process. The main objective of this study is to determine how data mining is implemented in determining the prediction of the birth process in pregnant women using the C4.5 algorithm. This research can provide knowledge about the combination of the Two Crows model and the C.45 algorithm to predict the risk status of childbirth in pregnant women. The C.45 algorithm is one of the most popular prediction techniques because it is easy for humans to interpret. The data analysis technique in this study uses the Two Crows model which is a development of the CRISP-DM model. The flow of the Two Crows model includes Understanding Business Problem, Building Data Mining Database, Data Explore, Prepare Data For Modeling, Building Model, and Evaluate Model. The data taken is examination data on pregnant women at the Health Center. Based on the results of the study, it was found that the highest root of the application of the C4.5 algorithm is in the height variable. The evaluation was carried out using a confusion matrix. From the evaluation results, it was found that the accuracy value reached 98.44%, the precision value reached 96%, and the recall value reached 100%.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call