Analisis Kelayakan Peminjaman Uang untuk Pembelian Properti Dipengaruhi oleh Status Perkawinan dan Jumlah Tanggungan Menggunakan Algoritma Naïve Bayes

Abstract

This study aims to analyse the feasibility of loan approval for property purchases influenced by marital status and the number of dependents using the Naïve Bayes algorithm. Data were collected from a bank and analyzed using Orange Data Mining software. The results show that the Naïve Bayes algorithm is effective in predicting loan feasibility with an accuracy rate of 80.4%. Other evaluation metrics such as F1 score, precision, and recall also demonstrate good performance, with values of 78.2%, 81.4%, and 80.4% respectively. Although there are some weaknesses in predicting both positive and negative classes with equal accuracy, overall, the Naïve Bayes method remains reliable for this purpose. The implementation of this algorithm using the Orange Data Mining toolkit facilitates the data analysis and visualisation process, providing a clear understanding of the factors influencing loan feasibility for borrowers.