Customer Churn Prediction Using Multiple Classification Mode…

Follow the guidelines in the attached Word document. The chosen dataset for this case study is the Telco Customer Churn dataset used in Case Study 1, and the models Nave Bayes and Decision Tree from Case Study 1 will be reused. Using Weka, run four classification models: Nave Bayes, Decision Tree, kNN, and one additional model (Random Forest, Logistic Regression, or SVM). Perform kNN tuning with different k values and hyperparameter tuning for the new model. Evaluate all models using Stratified 10-fold Cross-Validation and report results as mean standard deviation (or confidence interval). Provide confusion matrices, Accuracy, Precision, Recall, F1, ROC/AUC, and Lift/Gains, and compare the models using statistical tests (Friedman and Wilcoxon). Include screenshots from Weka and present results in tables and graphs in a 1013 page report, ending with a conclusion, one limitation, and one improvement idea.

Please include the (Turnitin) AI report with the final submission. Reports without it cannot be accepted.