Pakpahan, Januar and Susanti, Meilia Nur Indah and Wulandari, Dewi Arianti (2024) SISTEM REKOMENDASI PENGAJUAN KREDIT KONSUMTIF NASABAH BERBASIS DATA MINING DENGAN METODE NAÏVE BAYES. Diploma thesis, ITPLN.
202031163_Januar Pakpahan_Revisi_Skripsi_JANUAR Pakpahan.pdf
Restricted to Registered users only
Download (3MB)
Abstract
Di era teknologi modern yang mendorong pertumbuhan cepat industri perbankan, persaingan antar lembaga keuangan semakin meningkat, terutama dalam pemberian kredit, yang merupakan fungsi utama bank. Keputusan pemberian kredit sangat krusial karena berisiko terhadap kredit macet. Oleh karena itu, diperlukan sistem yang dapat memprediksi kelayakan kredit nasabah dengan tepat. Penelitian ini menggunakan metode Naive Bayes untuk mengembangkan sistem rekomendasi pengajuan kredit nasabah. Metode ini memanfaatkan probabilitas dari berbagai atribut nasabah, seperti umur, jenis kelamin, status pekerjaan, status perkawinan, tabungan, dan tujuan peminjaman, untuk menilai kelayakan kredit. Naive Bayes adalah algoritma klasifikasi yang mengkategorikan data berdasarkan probabilitas, dengan asumsi bahwa setiap fitur saling independen. Hasil penelitian menunjukkan bahwa sistem yang dikembangkan memiliki tingkat akurasi 70,50%, presisi 92,16%, recall 64,57%, dan F1-score 75,83%. Evaluasi kinerja dilakukan dengan menggunakan data uji untuk memastikan model dapat diandalkan dalam konteks nyata. Sistem ini diharapkan dapat membantu bank dalam membuat keputusan ditolak atau diterima terkait pengajuan kredit nasabah
In the modern technological era that drives the rapid growth of the banking industry, competition among financial institutions is intensifying, particularly in credit issuance, which is a core function of banks. Credit approval decisions are crucial due to the risk of non-performing loans. Therefore, a system is needed that can accurately predict customer creditworthiness. This study employs the Naive Bayes method to develop a credit application recommendation system. This method leverages the probability of various customer attributes, such as age, gender, employment status, marital status, savings, and loan purpose, to assess creditworthiness. Naive Bayes is a classification algorithm that categorizes data based on probability, assuming that each feature is independent of the others. The study results show that the developed system achieves an accuracy rate of 70.50%, a precision of 92.16%, a recall of 64.57%, and an F1-score of 75.83%. Performance evaluation was conducted using test data to ensure the model's reliability in real-world contexts. This system is expected to assist banks in making accept or reject decisions regarding customer credit applications.
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Sistem Rekomendasi,Kredit Macet,Pemberian Kredit,Klasifikasi,Data Mining,Naive Bayes,Probalitas,Crisp-DM Recommendation System, Non-performing Loans, Credit Approval, Classification, Data Mining, Naive Bayes, Probability, CRISP-DM |
| Subjects: | Skripsi Bidang Keilmuan > Teknik Informatika |
| Divisions: | Fakultas Telematika Energi > S1 Teknik Informatika |
| Depositing User: | Sudarman |
| Date Deposited: | 01 Oct 2025 06:25 |
| Last Modified: | 01 Oct 2025 06:25 |
| URI: | https://repository.itpln.ac.id/id/eprint/1657 |
