Prediksi Kelayakan Peminjam berdasarkan Data Pinjaman Menggunakan Credit Scoring

Radriyantami, Hayu Ajeng (2021) Prediksi Kelayakan Peminjam berdasarkan Data Pinjaman Menggunakan Credit Scoring. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 05111740000151-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (6MB) | Request a copy

Abstract

Selain bank, bermunculan lembaga atau perusahaan keuangan yang menawarkan layanan untuk memberikan pinjaman bagi masyarakat. Namun, tidak sedikit masyarakat yang mengalami gagal bayar. Akibatnya, lembaga keuangan mengalami kerugian karena telah salah memilih orang untuk diberikan pinjaman. Untuk mengatasi masalah tersebut, maka dilakukan prediksi kelayakan peminjam dengan Credit Scoring menggunakan pemodelan Logistic Regression. Sebelum melakukan pemodelan, dilakukan transformasi data oleh Weight of Evidence (WOE) dan menyeleksi fitur oleh Information Value (IV) dan feature selection. Model dievaluasi dengan AUC, ROC, dan K-Fold Cross Validation. Setelah itu, scorecard digunakan untuk memprediksi kelayakan peminjam. Hasil dari Tugas Akhir berupa prediksi kelayakan peminjam antara diterima atau ditolak. Berdasarkan eksperimen penyebab seseorang ditolak karena past default, debt-to-income terlalu tinggi, atau credit score terlalu kecil.
======================================================================================================
Besides banks, there are financial institutions or
companies that offer services to provide loans to the public.
However, not a few people who experience default. As a result,
financial institutions suffer losses because they have chosen the
wrong person to be given a loan. To overcome this problem, a
borrower's eligibility prediction is made with Credit Scoring using
Logistic Regression modeling. Prior to modeling, data
transformation is carried out by Weight of Evidence (WOE) and
selecting features by Information Value (IV) and feature selection.
The model was evaluated by AUC, ROC, and K-Fold Cross
Validation. After that, the scorecard is used to predict the eligibility
of the borrower. The results of the Final Project are in the form of
predictions of the borrower's eligibility between being accepted or
rejected. Based on experiments, the cause of someone being
rejected because of past default, debt-to-income is too high, or
credit score is too low.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Credit Scoring, Logistic Regression, Weight of Evidence, Information Value, feature selection, scorecard
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) T Technology > T Technology (General) > T174.5 Technology--Risk assessment. T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Hayu Ajeng Radriyantami
Date Deposited:	10 Aug 2021 04:51
Last Modified:	10 Aug 2021 04:51
URI:	http://repository.its.ac.id/id/eprint/85357

Actions (login required)

View Item