Prediksi Kelayakan Peminjaman Menggunakan Support Vector Machine (svm)

Wulandari, Febrika (2024) Prediksi Kelayakan Peminjaman Menggunakan Support Vector Machine (svm). Other thesis, Universitas Islam Riau.

Text 193510054.pdf - Submitted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

This research aims to apply the Support Vector Machine (SVM) method in predicting borrower eligibility as a solution to the problem of accumulating credit application data caused by a large number of applicants, making it difficult for employees in the credit granting consideration process. By using SVM, this research went through several stages, from reading the dataset, handling missing values, transforming categorical values into numeric, separating the dataset into features and targets, dividing training and test data, initializing and training the SVM model, to evaluating model performance by calculating prediction accuracy on test data. This research produces fast and accurate predictions of borrower eligibility. In this study, the results of model evaluation using the white box test showed an accuracy level of 0.75, with precision of 0.75, recall of 0.83, and F1- score of 0.78. This indicates that the model performs well in predicting borrower eligibility, with 15 of the 20 data samples tested predicted correctly.

Actions (login required)

View Item