Penerapan Algoritma Naive Bayes untuk Penentuan Resiko Kredit Kepemilikan Kendaraan Bermotor

Abstract

Credit companies provide rentals for the community to help obtain transportation vehicles such as motorbikes, but also require the approval received by companies such as bad payments in the payment of loans issued to companies that make money or financing. With the many credit risks, companies must be selective in choosing consumers who will be given credit so as not to make the company suffer losses. The method used in this study is the Naïve Bayes Algorithm and processed using Rapidminer Studio 5.3 software. The data used consisted of 55 training data and 10 test data. There are 9 variables used in this study, namely marital status, number of children, home ownership, housing conditions, employment, tenure, age, income, and down payment. The level of accuracy obtained from testing using Rapidminer Studio 5.3 is equal to 90%. By using the Naive Bayes algorithm in classifying consumers, it is hoped that it can assist companies in carrying out the mining process on the previous data and make the right decisions in determining new consumers from previous consumer data.