Negative Correlation Learning for Customer Churn Prediction: A Comparison Study

Table 1

Related work for churn prediction methods.

Author	Method	Description
Idris et al. [17]	GP	GP is applied with AdaBoost for churn prediction
Tsai and Lu [16]	ANN with BP	Applied as a hybrid approach in two stages (i.e., reduction and prediction)
Wang and Niu [14]	SVM	Least squares support vector machine (LS-SVM) is applied to establish a prediction model of credit card customer churn
Eastwood and Gabrys [31]	IBK	Authors apply simple -nearest neighbor algorithm on a nonsequential representation of sequential data for churn prediction
Kraljevíc and Gotovac [32]	Decision trees	Decision trees (DT) were applied and compared with ANN and logistic regression. DT results outperform other models
Verbraken et al. [33]	Naive Bayes	Number of Bayesian Network algorithms, ranging from the Naive Bayes classifier to General Bayesian Network classifiers are applied for churn prediction