DETEKSI CALON KREDITUR MOTOR DENGAN NAÏVE BAYES CLASSIFIER (STUDI KASUS: PT. FIF CABANG KUPANG)

Abstract

Prospective customers of PT. Federal International Finance (FIF) who applied for motor credits came from various backgrounds of work, residence and character. Therefore, given a large number of submission of motor credits every month as well as the various types of prospective customers, it is very necessary a system that is able to handle the problem of appraisal appropriateness applying for motor credits accurately. To obtain an accurate and precise classification value in order to produce a classification value with a good accuracy, the data used will be trained with the method of Naive Bayes Classifier by dividing the data using k = 4 to obtain the best model, resulting in a value with the result of accuracy calculated by Measured the comparison of the target class against the actual class, and the resulting measurement being 78% Precision, Recall 54%, with an overall accuracy of 68%. The system is expected to take decisions quickly and accurately in order to assist the Head Unit of credit in determining the decision of credit appropriately and quickly.