Problem: It is desired to develop a machine learning model that can predict customers who will leave the company. You are expected to perform the necessary data analysis and feature engineering steps before developing the model.
Solution process:
1-Exploratory Data Analysis
2-Feature Engineering
3-Modelling
4-Base model success rate:
- Base Model
- Accuracy: 0.7837
- Recall: 0.6333
- Precision: 0.4843
- F1: 0.5489
- Auc: 0.7282
5-Hyperparameter Optimization
6-Final Model
7-Feature Importance
All applied models and success rates are as follows:
******** LR ********
Accuracy: 0.8009
Auc: 0.8434
Recall: 0.5415
Precision: 0.6497
F1: 0.5904
******** KNN ********
Accuracy: 0.7698
Auc: 0.7802
Recall: 0.5158
Precision: 0.5747
F1: 0.5432
******** CART ********
Accuracy: 0.7363
Auc: 0.6666
Recall: 0.5142
Precision: 0.5044
F1: 0.5088
******** RF ********
Accuracy: 0.7899
Auc: 0.819
Recall: 0.4858
Precision: 0.6378
F1: 0.5513
******** SVM ********
Accuracy: 0.7931
Auc: 0.7827
Recall: 0.4596
Precision: 0.659
F1: 0.5407
******** XGB ********
Accuracy: 0.7872
Auc: 0.8238
Recall: 0.5212
Precision: 0.6185
F1: 0.5652
******** LightGBM ********
Accuracy: 0.7954
Auc: 0.8366
Recall: 0.5217
Precision: 0.642
F1: 0.5752 '''
#Feature Importance ia as follows:
