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

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

• By Sopian, Abu; Wiyatno, Agus; Riyandi, Albert;
• Jurnal Informatika
• 158
• 88-95

Credit  may  be  meant  money  provision  or collection that  can  be  equavalent  with  that,  based oncredit approval or loan agreement between bank and  other  party  who  oblige  lender  to  pay  off  the debtafter   specific   terms   period   with   interest expenses.  Commercial  Bank  is  a  bank  that  operate itsbusiness   in   conventional   and   or   based   on syariah  principle  which  is  in  operation  provide  in and outpayment service. In this business operation, commercial bank provides loan/credit facility to thecustomer  in  Rupiah  and foreign  currency. Working capital  credit  is  a  creditused  to  finnance  workingcapital 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  capitalcredit approval  providedby  commercial  bank  need  topredict  because  it  has increased     of     credit     provision     provided     bycommercial  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   thisresearch   will   conducted working   capital   creditvalue   approval   prediction   will   be   provided   by commercialbank   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 accuracyresult :68,8% and RMSE : 11928,594and  the  result  acquired  using  artificial neutral  network  algorithmproviding  the  accuracy result  :84,7%  and  RMSE  : 5806,350.  This  result shows   that   the   bestperformance   for working capital    creditvalue    approval    provided    by commercial   bank   is   artificial   neutralnetwork algorithm

• Katakunci Credit, Working Capital Credit, Support Vector Machine, Artificial Neural Network, Commercial Bank
• URL http://journal.thamrin.ac.id/index.php/jtik/article/view/224/188
• Penerbit ,

REFERENSI