Kaggle Notebook: https://www.kaggle.com/code/shasan7/bank-churn
Used scikit-learn's machine learning models for prediction. Logistic Regression, Random Forest, Support Vector Machine, XGBoost, CatBoost, Ridge Classifier and SGD Classifiers were run.
The XGBoost attained an accuracy of 97% with an F1-Score of 0.94, performing best among the machine learning algorithms. We trained it with 100 estimators using the "logloss" evaluation metric, and 'balanced' value was passed to the class_weights parameter to provide attention to the minority class, which is crucial for a dataset like ours with huge class imbalance.
Obtained the following results:
Classifiers: LogisticRegression has a score of 85.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.96 0.86 0.91 1701
1 0.52 0.81 0.63 325
accuracy 0.85 2026
macro avg 0.74 0.83 0.77 2026
weighted avg 0.89 0.85 0.86 2026
Confusion Matrix:
[[1461 240]
[ 63 262]]
Classifiers: RandomForestClassifier has a score of 95.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.96 0.99 0.97 1701
1 0.93 0.77 0.84 325
accuracy 0.95 2026
macro avg 0.94 0.88 0.91 2026
weighted avg 0.95 0.95 0.95 2026
Confusion Matrix:
[[1683 18]
[ 76 249]]
Classifiers: RidgeClassifier has a score of 84.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.96 0.85 0.90 1701
1 0.51 0.81 0.63 325
accuracy 0.84 2026
macro avg 0.73 0.83 0.76 2026
weighted avg 0.89 0.84 0.86 2026
Confusion Matrix:
[[1448 253]
[ 62 263]]
Classifiers: SGDClassifier has a score of 84.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.96 0.85 0.90 1701
1 0.50 0.80 0.62 325
accuracy 0.84 2026
macro avg 0.73 0.83 0.76 2026
weighted avg 0.88 0.84 0.86 2026
Confusion Matrix:
[[1446 255]
[ 65 260]]
Classifiers: SVC has a score of 93.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.97 0.94 0.95 1701
1 0.73 0.84 0.78 325
accuracy 0.93 2026
macro avg 0.85 0.89 0.87 2026
weighted avg 0.93 0.93 0.93 2026
Confusion Matrix:
[[1601 100]
[ 51 274]]
Classifiers: CatBoostClassifier has a score of 96.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.99 0.97 0.98 1701
1 0.85 0.93 0.89 325
accuracy 0.96 2026
macro avg 0.92 0.95 0.93 2026
weighted avg 0.96 0.96 0.96 2026
Confusion Matrix:
[[1647 54]
[ 23 302]]
Classifiers: XGBClassifier has a score of 97.0 % accuracy score.
Classification Report:
precision recall f1-score support
0 0.97 0.99 0.98 1701
1 0.93 0.86 0.90 325
accuracy 0.97 2026
macro avg 0.95 0.92 0.94 2026
weighted avg 0.97 0.97 0.97 2026
Confusion Matrix:
[[1681 20]
[ 45 280]]
Feature importance for the XGBoost Classifier:
