<![CDATA[ABSTRAKKeberadaan lembaga perbankan sangat penting bagi kehidupan masyarakat, terutama digunakan untuk menghimpun dana baik dalam bentuk deposito, giro, tabungan dan lainnya. Disamping hal itu lembaga perbankan juga dapat berperan sebagai penyalur bantuan dana dalam bentuk kredit kepada masyarakat dan badan usaha yang memerlukan. Permasalahan dalam penyeluran kredit menyebabkan kredit macet dari nasabah sehingga menyebabkan kerugian pada bank tersebut. Penilaian kredit merupakan salah satu tahapan penting yang harus dilakukan oleh pihak bank sebelum kredit diberikan kepada pemohon kredit. Proses penilaian kredit tergolong kedalam permasalahan semi terstruktur yang cukup kompleks, oleh sebab itu dibutuhkan pemodelan sistem untuk memprediksi kelayakan terhadap pengajuan kredit. Pemodelan yang dibangung dalam penelitian ini menggunakan algoritma K-Nearest Neighbor Classification dengan menilai calon debitur dari data training yang digunakan. Berdasarkan penelitian yang sudah dilakukan, algoritma K-Nearest Neighbor Classification dapat dimodelkan dalam penilaian kredit perbankan. Hasil pengujian model menunjukkan akurasi ketepatan perhitungan sebesar 83%.Kata kunci: perbankan, kredit, debitur, K-Nearest Neighbor ClassificationABSTRACTThe existence of banking institutions is very important for people's lives, mainly used to raise funds in the form of deposits, demand deposits, savings and others. Besides that, banking institutions can also play a role as distributors of financial assistance in the form of credit to the public and business entities that need it. Problems in credit disbursement cause bad loans from customers, causing losses to the bank. Credit assessment is one of the important stages that must be carried out by the bank before credit is given to the credit applicant. The credit appraisal process is classified as a semi-structured problem that is quite complex, therefore system modeling is needed to predict the feasibility of applying for credit. The modeling built in this study uses the K-Nearest Neighbor Classification algorithm by assessing prospective borrowers from the training data used. Based on the research that has been done, the K-Nearest Neighbor Classification algorithm can be modeled in a bank credit assessment. The results of the model testing showed the accuracy of the calculation accuracy was 83%.Keywords: banking, credit, debtor, K-Nearest Neighbor Classification]]>
