Prediction of default payment of credit card clients using Data Mining Techniques

Abstract

Recent studies showed that poor people had trouble with their financial decisions. In order to prevent these financial complications including decisions which are easier and more frequent in the unintentional failure for paying monthly credit card balances, we proposed a data mining-based failure prevention system from the view of risk management. This research realized on customers' default payments for the accurate prediction of the probability of default payment. Because the class of default dataset is imbalanced dataset, this study presents the Synthetic Minority Over-Sampling Technique (SMOTE) to deal with the imbalanced dataset. By utilizing SMOTE method, the predicting model produced by Random Forest has the best accuracy and performance with low error rate. Consequently, among the seven data mining algorithms, Random Forest is a good alternative to precisely predict the default payment. The proposed Random Forest model classifies Default of Credit Card Clients in the test data by predicting the target variable 89.01% correctly. This effect also resulted in an improvement of ROC area (AUC=0.947), and F-measure (0.89) of Random Forest, which was higher than other classifiers.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 14 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.