Abstract:
Credit risk assessment for secured loans is an important operation in
banking systems as financing loans gives banking institutions a substantial profit. It
is also important that banks finance right lenders who pay the loans on schedule and
on time. However, there are always non-performing loans that may risk the bank
institutions. This paper aims to identify factors which are necessary for bank
institutions to assess credit application. By aiming on the reduction of number of nonperforming
loans, current decision criteria on credit risk assessment are evaluated.
Subsequently, a decision tree model is proposed by applying data mining
methodology.
The case study of the assessment is current credit assessment at PT BPR X
in Bali. Based on last 10 year data, there are 1082 lenders or 11.99% which have
non-performing loans and those are identified as bad loan cases. This made PT BPR
X may be categorized as a poorly performing bank.
The methodology of data mining is used because it cat indicate whether the
request of lenders can be classified as performing or non-performing loans risk. This
model is designed in accordance with the standards process in the application of
data mining and using Clementine software, a software tool for data mining, then a
new decision tree model is generated. This model suggests that new criteria in
analyzing the loan application. The evaluation results shot that if this model is
applied to unseen data, can reduce the percentage of non-performing loans below
5% or limit a bank can be classified as healthy banks by BI.