Komparasi Metode Algoritma Data Mining pada Prediksi Uji Kelayakan Credit Approval pada Calon Nasabah Kredit Perbankan

Abstract

ABSTRAKSI D i e r a s a a t i n i p e m a n f a a t a n d a t a m i n i n g d a l a m m e n e t u k a n k e b e r h a s i l a n p e n i n g k a t k a n l a y a n a n p e r b a n k a n s a n g a t l a h e f e k t i f , s a l a h s a t u n y a p e m b e r i a n k r e d i t k e p a d a c u s t a m e r b a n k , m e n e n t u k a n c r e d i t a p p r o v a l m e m e r l u k a n s i s t e m y a n g a k a n d i g u n a k a n u n t u k m e n j a l a n k a n p r o s e s p e n g a j u a n k r e d i t s e r t a d i d u k u n g d e n g a n k e r c e r m a t a n d a l a m m e m i l a h c a l o n n a s a b a h k r e d i t y a n g b a i k s e h i n g g a d a p a t m e m i n i m a l i s i r t e r j a d i n y a k r e d i t m a c e t y a n g t i d a k d i n g i n k a n. U n t u k m e n d u k u n g h a s i l d a r i t i n g k a t k e b e r h a s i l a n m a r k e t i n g d a l a m p e r a n n y a u n t u k m e m a s a r k a n p r o d u k l a y a n a n p e r b a n k a n y a n g p r o s e s n y a m e m b u t u h k a n d a t a c a l o n n a s a b a h i n i , m a k a d u k u n g a n d a t a m i n i n g s a n g a t be r p e r a n p e n t i n g d a l a m k l a s i f i k a s i c a l o n n a s a b a h b a n k y a n g a k a n m e n g a m bi l k r e d i t d i b a n k. B e r d a s a r k a n p e m e t a a n p e n e l i t i a n m e n g e n a i d u k u n g a n d a t a m i n i n g p a d a c a l o n n a s a b a h d i d a p a t a d a a l g o r i t m a k l a s i f i k a s i y a n g s e r i n g d i g u n a k a n u n t u k k l a s i f i k a s i c a l o n n a s a b a h a n t a r a l a i n N e u r a l N e t w o r k , N a i v e B a y e s d a n K-N N d a l a m p r e d i k s i k e b e r h a s i l a n m a r k e t i n g d a l a m m e n e n t u k a n k e l a y a k a n d a r i n a s a b a h p e m i n j a m k r e d i t b a n k d a r i u j i c o ba y a n g d i l a k u k a n m a k a a l g o r i t m a N e u r a l N e t w o r k l a h y a n g l e bi h a k u r a t d e n g a n a k u r a s i 9 0 , 7 1 % d e n g a n n i l a i A U C 0. 8 8 0 , h a l i n i d a p a t m e n j a d i p e r b a n d i n g a n d a t a m i n i n g k l a s i f i k a s i M e l i h a t n i l a i A U C d a r i k e t i g a m e t o d e t e r s e b u t y a i t u N N , N a i ve B a y e s d a n K-N N , m a k a 3 a l g o r i t m a t e r s e bu t t e r m a s u k k e l o m p o k k l a s i f i k a s i b a i k k a r e n a n i l a i A U C-n y a a n t a r a 0. 8 0-1. 0 0. ABSTRACT In the current era, the use of data mining in determining the success of improving banking services is very effective, one of which is lending to bank custers, determining credit approval requires a system that will be used to run the credit application process and supported by careful selection of prospective credit customers so that they can minimize the occurrence of bad credit. To support the results of the success rate of marketing in its role in marketing banking service products whose processes require prospective customer data, data mining support plays an important role in the classification of prospective bank customers who will take credit at the bank. classification algorithm that is often used for the classification of prospective customers, including NN, NB and K-NN in predicting marketing success in determining the feasibility of bank loan borrowers then the NN algorithm is more accurate with an accuracy of 90.71% with an AUC 0.880 value, this data mining classification comparison Seeing the AUC value of the 3 methods namely NN, NB and K-NN, then the 3 algorithms are included in the group classification is good because the AUC value is between 0.80-1.00.