Entering the era of the Covid-19 pandemic, lending capital (credit) to KOPPAS has increased, causing a lot of data and slowing down the credit disbursement process. Currently, KOPPAS has not implemented a support system that can provide consideration for credit granting decisions. Therefore, in this study, researchers will discuss the design and development of a decision support application system for granting credit using the Support Vector Machine (SVM) method, to speed up the credit granting process, prevent the level of subjectivity in granting credit, and to minimize the occurrence of bad loans. resulting from errors in the decision-making process. The development of this system uses modeling techniques with data classification processes and the CRISP-DM development method. In this case, the Support Vector Machine method classifies data into 2 classes, namely feasible and infeasible classes. The model development process uses the Radial Basis Function (RBF) kernel parameter by exploring the use of C (cost) and y (gamma) parameter values. This development model uses 200 data with the attributes used in the form of the loan amount, term, collateral value, guarantor agreement, occupation, and loan history. The use of these attributes and the amount of data can produce quite good performance with the highest accuracy rate of 85% using C = 1 and y = 0.005 and the distribution of data is 80% training data and 20% testing data. With the performance of this model, the system can classify creditworthiness.
Copyrights © 2023