Implementation of Data Mining to Classify Potential Customers Using the C5.0 Algorithm

Rizki, Muhammad
Maghfirah, Cintya Nil
Norhiza, Fitra Lestari
Nofirza, Nofirza
Lubis, Fitriani Surayya

Jurnal Teknik Industri : Jurnal Hasil Penelitian dan Karya Ilmiah dalam Bidang Teknik Industri
Vol 11, No 1 (2025): JUNI 2025

PT Pegadaian, as listed on its official website, is a fast-growing financial company. One of its key challenges is late installment payments, which can lead to financial losses. Using pawn customer data from 2013 to 2021, this study found that out of 534 customers, 68 were late in paying installments, and 10 did not pay. To address this issue, this research applies customer classification to identify borrowers who are more likely to pay on time. The classification model is developed using data mining with the C5.0 algorithm to generate decision-tree rules. Prior to modeling, the dataset is processed through the Knowledge Discovery in Databases (KDD) stages, including data selection, cleaning, and transformation. The proposed model produces 26 classification rules and achieves an accuracy of 87.04%. All data processing, modeling, and validation are conducted using RapidMiner Studio. Keywords: Classification, Decision Tree, C5.0 Algorithm, Data Mining