Customer Churn Modeling via the Grey Wolf Optimizer and Ensemble Neural Networks

Table 3

Comparing the performance of the proposed system and other methods in different evaluation criteria.

No.	Methods	F_measure	Recall	Precision	Accuracy
1	Naive Bayes	72.3	83.9	63.6	73.0
2	Generalized linear model	71.6	72.8	70.4	75.7
3	Logistic regression	71.6	72.9	70.40	75.7
4	Deep learning	72.6	80.9	66.0	74.3
5	Decision tree	70.7	84.6	61.10	70.7
6	Random forest	70.5	70.5	70.6	75.2
7	Gradient boosted tree	76.2	79.6	73.1	79.1
8	The proposed method	87.64	91.08	84.45	80.84