Perbandingan Algoritma Naive Bayes dan Naive Bayes Berbasis PSO untuk Analisis Kredit pada PT. BPR Syariah Paokmotong

Abstract

According to Law No.21 of 2008, it is stated that Islamic banks are business entities that collect funds from the public in the form of deposits and channel to the public in the form of loans and / or other forms in order to improve people's lives. Credit is the provision of money or bills that can be equated with it, based on a loan and loan agreement between the bank and another party which requires the borrower to repay the debt after a certain period of time with interest (Banking Law No.10 of 1998). In analyzing a credit, sometimes an analysis performs an inaccurate analysis, so there are some customers who are less able to pay credit installments, resulting in substandard loans and even defaults. From the above problems, researchers conducted a credit analysis using computerized techniques by utilizing RapidMiner software in processing data. The right data processing technique to use is classification. One method of classification of data mining is the Naive Bayes algorithm. The researcher used weighting by applying Particle Swarm Optimization (PSO) for attribute selection to improve the accuracy of Naive Bayes. After testing with the Naive Bayes algorithm model of 71.00%, the Naive Bayes algorithm based on particle swarm optimization results in a higher accuracy value of 88.51% compared to the Naive Bayes algorithm model. From these results the difference between the two models is 17.51%. DOI : 10.29408/jit.v2i2.1460