Analisis Statistik Pembayaran Kredit Menggunakan Metode Random Forrest Classfier Dan LightGBM

Akni Widiyastuti (Fakultas Ilmu Komputer, Institut Teknologi Bisnis dan Bahasa Dian Cipta Cendikia)
Chairani Chairani (Fakultas Ilmu Komputer, Informatics & Business Institute Darmajaya, Lampung)
Kurniawati Kurniawati (Fakultas Ilmu Komputer, Institut Teknologi Bisnis dan Bahasa Dian Cipta Cendikia)
Siti Aminah (Fakultas Ilmu Komputer, Institut Teknologi Bisnis dan Bahasa Dian Cipta Cendikia)
Muhamad Abror (Fakultas Ilmu Komputer, Informatics & Business Institute Darmajaya, Lampun)

JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi)
Vol 6, No 3 (2023): JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi)

Banking also takes part in marketing technology based on financial services such as the following: Banking, Credit and Loans, Financial Advisory, and Consumer Convenience. By automating most of the banking processes and adding speed, transparency, and security to them, Fintech applications also indirectly contribute to a seamless experience for customers. It is known that one of the banks in Indonesia has also implemented a fintech system, one of which is in terms of processing credit payments (online credit). Some people have applied for credit at the bank by making payments digitally or online. Banks have collected data from their customers who have made credit online. These data are collected to obtain information that helps banks see the status of credit payments. In the research results, the accuracy value was obtained by using the Random Forrest Classifier and LightGBM method comparisons, the values obtained from the AUC for the validation set were around 65% and 80%.