Komparasi Algoritma Support Vector Machine Dengan Naïve Bayes Untuk Analisis Kelayakan Pemberian Kredit Usaha Mikro

Astofa, Aniq

INTEGER: Journal of Information Technology
Vol 6, No 2: September 2021

Credit has a high risk of credit congestion, this is due to the accidental factor due to the disaster experienced by the debtor so that the credit provided does not increase the income of the debtor, in addition to the existence of bad faith of the debtor by not fulfilling the obligation as it should, the data technique mining using Particle Swarm Optimization-based vector-based support method with Naive bayes. Support Vector Machine method has an accuracy of 50.70%. The second experiment conducted using Particle Swarm Optimization's Support Vector Machine method has an accuracy value of 85.92% and Compared with al-goritma or naive bayes method the accuracy value of 91.16% .with Rapidminer software