This project demonstrates a full end-to-end process from data cleaning, visualization, feature engineering, to model training, hyperparameter tuning, and evaluation, ultimately predicting customer churn.

1. Data Loading and Preprocessing

• Dataset: The Customer-Churn.csv file is loaded into a pandas DataFrame.
• Data Cleaning: The TotalCharges column, initially detected as an object type, is converted to a numeric type. Missing values are handled by dropping rows with null values.
• Feature Adjustment: The customerID column is dropped as it doesn’t contribute to predictions. The PaymentMethod column is modified to remove the term "automatic" from entries.

2. Data Visualization

• Churn Distribution: A bar plot shows the proportion of customers who have churned vs. those who haven’t, visualizing the class imbalance.
• Demographic Information: A stacked bar plot visualizes how demographic variables (gender, senior citizen status, partner, dependents) relate to churn.
• Customer Account and Services Information: Similar stacked plots explore variables like contract type, payment method, internet services, and support options.

3. Feature Engineering

• Label Encoding: Binary categorical features such as gender, Partner, PhoneService, and Churn are label-encoded (i.e., converted to numerical values).
• One-Hot Encoding: Categorical variables with more than two levels (e.g., InternetService, PaymentMethod, Contract) are transformed using one-hot encoding.
• Normalization: Min-max normalization is applied to continuous variables like tenure, MonthlyCharges, and TotalCharges to scale them to a range of [0,1], improving model performance.

4. Model Training and Evaluation

• Train-Test Split: The dataset is split into training (80%) and testing (20%) sets using train_test_split.
• Model Selection: Six machine learning models are evaluated:
  • Dummy Classifier
  • k-Nearest Neighbors
  • Logistic Regression
  • Support Vector Machine (SVM)
  • Random Forest Classifier
  • Gradient Boosting Classifier
• Accuracy Measurement: Each model is trained and evaluated on the test set using accuracy as the performance metric.

5. Hyperparameter Tuning

• Gradient Boosting Tuning: RandomizedSearchCV is used to optimize the hyperparameters of the Gradient Boosting Classifier, searching through various combinations of estimators, depth, features, and sample splits.
• Best Parameters: The best-performing combination of hyperparameters is identified and displayed.

6. Confusion Matrix

• The confusion matrix is generated to further evaluate the performance of the optimized Gradient Boosting model by showing the number of true positives, true negatives, false positives, and false negatives.