Komparasi Algoritma Naïve Bayes Dan Support Vector Machine (SVM) Untuk Klasifikasi Kelayakan Pemberian Pinjaman

Abstract

Cooperatives are social organizations or economic bodies that have a very important role in the growth, development of economic potential and community success. One of the cooperative activities is the provision of credit or loans to community members. Cooperative credit is one of the most important banking activities and serves to provide credit to the community. In practice, errors often arise due to inaccurate credit analysis, or the behavior of the customers themselves. The purpose of this research is to compare the accuracy results between the Naive Bayes algorithm and Support Vector Machine (SVM), where the best accuracy results can later be used as a reference to determine the profitability of lending. The attributes used in this study consist of 11 attributes, namely: Gender, marital status, occupation, relatives, nominal income, income criteria, loan amount, loan term, interest rate, installments and class as income characteristics. The dataset used in this study includes 166 members of the Daru Nahdla Capita Shari'ah cooperative. The results of testing the naive bayes algorithm after dividing the data five times, dividing the data set 70% as test data and 30% as training data, obtained a precision value of 97.00%, recall 100.00%, F1 score 99.00%. and accuracy 98.00%. Thus, the Naive Bayesian algorithm is an algorithm that shows accurate classification and prediction