Komparasi Algoritma Support Vector Machines dengan Algoritma Artificial Neural Network untuk Memprediksi Nilai Persetujuan Kredit Modal Kerja yang Diberikan Bank Umum

Abstract

Credit may be meant money provision or
collection that can be equavalent with that, based
on credit approval or loan agreement between bank
and other party who oblige lender to pay off the
debt after specific terms period with interest
expenses. Commercial Bank is a bank that operate
its business in conventional and or based on
syariah principle which is in operation provide in
and out payment service. In this business operation,
commercial bank provides loan/credit facility to the
customer in Rupiah and foreign currency. Working
capital credit is a credit used to finnance working
capital purposes are depleted in one or several time
the production. For example: to buy raw material,
salary, rent a building, purchase merchandise and
so forth. Working capital credit approval provided
by commercial bank need to predict because it has
increased of credit provision provided by
commercial bank that can be used as measurement
of economic growth and country stability or as
measurement of economic growth indicator from
monetary sector by Bank of Indonesia. In this
research will conducted working capital credit
value approval prediction will be provided by
commercial bank using support vector machine
algorithm that is compared with artificial neutral
network algorithm. From the result of testing on
support vector machine algorithm using kernel dot
providing the accuracy result : 68,8% and RMSE :
11928,594 and the result acquired using artificial
neutral network algorithm providing the accuracy
result : 84,7% and RMSE : 5806,350. This result
shows that the best performance for working
capital credit value approval provided by
commercial bank is artificial neutral network
algorithm.