<![CDATA[Adira Finance is one of the companies that runs in financing sector. Before providing financing to consumer, it is required to complete a series of processes to screen prospective consumer who are eligible for financing. However, an integral automated process is not available so far that is needed to improve the efficiency of the processing of financing approvals. One of them is determining the eligibility of a prospective consumer to be given financing by considering his tendency to make late payments. In this study, the tendency of a prospective consumer to make late payments is predicted using the random forest method by using financing history data and payment transactions for used motorcycles from the year 2017 to 2019. Stages to predict late pay tendencies begin with a series of data preprocessing processes, training data using a random forest algorithm to produce the most fit model, then the model is tested to obtain predictive results and the level of precision using a confusion matrix. Accuracy measurement on the prediction results get an average score of 88%. The output of this research is a prediction system that displays dashboard page, uploading new data, and prediction results. For testing user acceptance of the system, User Acceptance Testing is used and results an average of 87,5% which means that the user accepts the system that has been developed.]]>
Copyrights © 2020
