This model involves identifying at-risk customers who are likely to cancel their subscriptions or close/abandon their accounts. Various Classification models have been used and their hyperparameters tuned using GridSearchCV to obtain the most accurate model possible
Techniques used:
Machine learning modeling
Hyperparameter Tuning
Algorithms used:
1.Logistic Regression
2.SVM Classifier
3.K Nearest Neighbors (KNN)
4.Random Forest Classifier
5.Naive Bayes
6.AdaBoost
Model Evaluation Methods used:
1.Accuracy Score
2.ROC AUC Score
3.Confusion Matrix
4.Classification Report
Packages and Toold required:
1.Pandas
2.Matplotlib
3.Seaborn
4.Scikit-Learn
5.Google Colab