A Provisional Study of Data Mining Classification Algorithms in Predicting Credit Card Defaulters Using Weka Tools

Data mining is a field of study that seeks to draw out significant patterns from large amounts of data. This study explores the classification algorithms for spotting new trends in massive volumes of credit card data. This work covers the performance analysis of various data mining techniques, namely, Naïve Bayes, J48, and Support Vector Machine (SVM), in predicting credit card defaulters. The dataset is gathered from Kaggle, with 25 attributes and 30,000 instances used to assess the performance of algorithms. Additionally, each classification algorithm's effect on feature selection is mentioned. A variety of data mining classification approaches is used to compare the models in this study. The experiment finds J48 and SVM as the best algorithms at 81.56% and 80.92%, respectively.