Evaluating Model Performance
Calculate the model’s accuracy on both the training and testing datasets.
Training Accuracy
training_accuracy = accuracy_score(
Y_train,
training_predictions
)
print(training_accuracy)
Testing Accuracy
testing_accuracy = accuracy_score(
Y_test,
testing_predictions
)
print(testing_accuracy)
Precision Score
precision = precision_score(
Y_test,
testing_predictions
)
print(precision)
Recall Score
recall = recall_score(
Y_test,
testing_predictions
)
print(recall)
F1 Score
f1 = f1_score(
Y_test,
testing_predictions
)
print(f1)
ROC-AUC Score
roc = roc_auc_score(
Y_test,
testing_predictions
)
print(roc)
Explanation
Accuracy measures the proportion of correctly classified transactions.
However, because the dataset is highly imbalanced, additional evaluation metrics such as precision, recall, and F1-score should also be considered.
Credit Card Fraud Detection for Beginners using Data Science
VA










