<![CDATA[Non-performing loans is a problem that can have an impact on the running of cooperative business activities such as income from interest on loans and late or reduced capital. This is also a problem for Koperasi Serba Usaha (KSU) Surya Abadi, which in making decisions on loan applications still uses intuition. KSU Surya Abadi also still uses survey methods which require time and money. Therefore a system is needed that can provide recommendations or support decisions that can predict earlier related to non-performing loans on loan applications. Data mining is a process that can be used to make non-performing loans classifications. By using the C5.0 algorithm which is one of the classification algorithms, it is used to predict bad loans in loan applications that can produce a rule in the form of a decision tree. From the results of the evaluation and validation of the algorithm using the confusion matrix, an accuracy of 84% is obtained. Then, the resulting AUC value is based on the ROC curve of 0.836. To test the usability of the system using the System Usability Scale (SUS) the resulting value of 81.67. The resulting system is a dashboard visualization that contains several graphs to load time series, percentages, and trends of total loan submissions, form for predicting loan applications, form for entering new datasets, displaying accuracy and decision trees, user manager, and prediction attributes manager.]]>
Copyrights © 2020
