One-Step Dynamic Classifier Ensemble Model for Customer Value Segmentation with Missing Values

Table 6

Comparison of churn prediction performance in “China churn” dataset.

Model	Total accuracy	AUC	Type I accuracy	Type II accuracy
LMF	0.8124	0.8480	0.7364	0.8201
ODCEM	0.9313	0.9057	0.8539	0.9391
KI-SVM	0.8538	0.8747	0.8229	0.8570
MS-SVM	0.8638	0.8793	0.8022	0.8701
EM-SVM	0.8954	0.8820	0.7899	0.9061
RI-SVM	0.8280	0.8224	0.7088	0.8401

Note: the bold-face in Table 6 shows the maximum of each column.