- Communications Faculty of Sciences University Ankara Series A1 Mathematics and Statistics
- Volume:70 Issue:2
- Predicting credit card customer churn using support vector machine based on Bayesian optimization
Predicting credit card customer churn using support vector machine based on Bayesian optimization
Authors : Kamil Demirberk ÜNLÜ
Pages : 827-836
Doi:10.31801/cfsuasmas.899206
View : 51 | Download : 15
Publication Date : 2021-12-31
Article Type : Research Paper
Abstract :In this study, we have employed a hybrid machine learning algorithm to predict customer credit card churn. The proposed model is Support Vector Machine insert ignore into journalissuearticles values(SVM); with Bayesian Optimization insert ignore into journalissuearticles values(BO);. BO is used to optimize the hyper-parameters of the SVM. Four different kernels are utilized. The hyper-parameters of the utilized kernels are calculated by the BO. The prediction power of the proposed models are compared by four different evaluation metrics. Used metrics are accuracy, precision, recall and F 1 -score. According to each metrics linear kernel has the highest performance. It has accuracy of %91. The worst performance achieved by sigmoid kernel which has accuracy of %84.Keywords : Churn analysis, support vector machine, machine learning, hyper parameter optimization
ORIGINAL ARTICLE URL