Text

PREDIKSI TINGKAT RISIKO KREDIT DENGAN RANDOM OVER-UNDER SAMPLING PADA METODE ENSEMBLE MENGGUNAKAN ALGORITMA DECISION TREE ID3, RANDOM FOREST DAN REGRESI LOGISTIK BINER

---

Penilaian

---

Credit-granting activities are included in business activities that have a high risk and affect the sustainability of the company as well as other financial institutions. In credit activities, non-performing loans often occur due to the failure to repay a number of loans in accordance with the agreed time. The problem of providing credit can be overcome, one of which is by identifying and predicting prospective customers before giving credit. Datasets used to predict sometimes have class imbalance problems. This problem is usually solved by resampling method. Therefore, this research was conducted with the aim of predicting the level of credit risk by implementing Random Over-Under Sampling in the Ensemble method using Decision Tree ID3, Random Forest, and Binary Logistics Regression. The data used is a dataset of credit card approval UCI Repository. The results showed that the Ensemble method has a better overall classification effectiveness level than others, as seen from the higher accuracy, precision, and fscore values, while the better classification effectiveness level in the form of recall is Binary Logistics Regression. Prediction classification using decision tree resulted in accuracy, precision and recall of 77.79%, 49.82, 45.95%, 47.76%, respectively. Prediction classification using random forest resulted in accuracy, precision and recall of 78.10%, 50.55%, 45.31%, 47.76%, respectively. Prediction classification using binary logistic regression resulted in accuracy, precision and recall of 74.16%, 42.66%, 48.90%, 45.55%, respectively. Prediction classification using ensemble majority vote resulted in accuracy, precision and recall of 78.22%, 50.86%, 45.54%, 48.03%, respectively. Keywords: Credit Risk, Ensemble, Decision Tree, Random Forest, Binary Logistics Regression

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2207001781	T72258	T722582022	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T722582022

Publisher

Inderalaya : Jurusan Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya., 2022

Collation

xiv, 92 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

518.107

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Algoritma
Jurusan Matematika

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment