<p>Mendeteksi kualitas kredit sejak dini merupakan satu tahapan penting yang wajib dilakukan oleh koperasi simpan pinjam guna meminimalisir adanya risiko kredit. Dalam penelitian ini, kami menggunakan tiga metode klasifikasi yaitu SVM, <em>Neural Network</em>, dan <em>Naïve Bayes</em> untuk menemukan metode dengan performa yang paling baik dan optimal pada kasus pendeteksian kualitas kredit di koperasi simpan pinjam. Proses yang dilakukan adalah dengan mengimplementasikan data hasil <em>pre processing</em> menggunakan algoritme SVM, <em>Neural Network</em>, dan <em>Naïve Bayes</em> dengan proses evaluasi menggunakan <em>5-fold cross validation</em>. Hasil yang didapatkan adalah metode <em>Neural Network</em> menjadi metode dengan performa paling baik. Rerata tingkat akurasi yang dihasilkan sebesar 86,81%, rerata <em>precision</em> sebesar 0,8194, rerata <em>recall</em> sebesar 0,8236, dan rerata nilai AUC sebesar 0,9158. Namun, waktu eksekusi yang dihasilkan algoritme <em>Neural Network</em> menjadikan algoritme ini sebagai algoritme paling lambat dibandingkan dengan dua metode lain. Nilai rerata waktu eksekusi dari metode <em>Neural Network</em> sebesar 3,058 detik, jauh lebih lama dibandingkan dua algoritme lain yang hanya berkisar pada nilai 0 – 1 detik.</p><p> <strong><em>Abstract</em></strong></p><p><em>Detecting credit quality at the early stage is an important step that must be done by koperasi simpan pinjam in order to minimize the credit risk. In this research, we use three classification methods i.e. SVM, Neural Network, and Naïve Bayes to find the best performance and optimal method to be used in credit quality detection for koperasi simpan pinjam. The process conducted by implementing pre-processing data using an SVM, Neural Network, and Naïve Bayes algorithm with the evaluation process using 5-fold cross validation. As the result, The Neural Network method was the best performing method. The average level of accuracy produced was 86.81%, mean precision was 0.8194, average recall was 0.8236, and the average AUC value was 0.9158. However, the execution time generated by the Neural Network algorithm made this algorithm the slowest algorithm compared to the other two methods. The average execution time of the Neural Network method was 3.058 seconds, longer than the other two algorithms which only range from 0 - 1 second.</em></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.