Telco Customer Churn Prediction Overview This project aims to predict customer churn for a telecommunications company using machine learning models. By identifying customers likely to churn, the company can take proactive measures to retain them and improve business performance.

Dataset The dataset used for this project is the "Telco Customer Churn" dataset from Kaggle. It contains 7,043 entries with 21 columns, including customer demographics, account information, and service usage details.

Project Steps

1. Data Exploration and Cleaning Converted TotalCharges to numeric and handled missing values. Dropped the customerID column. Performed one-hot encoding on categorical features. Created new features: tenure groups and average monthly charges.
2. Feature Selection Used Recursive Feature Elimination (RFE) with a RandomForestClassifier to select the top 10 significant features: tenure MonthlyCharges TotalCharges avg_monthly_charges gender_Male InternetService_Fiber optic Contract_One year Contract_Two year PaperlessBilling_Yes PaymentMethod_Electronic check
3. Model Building Implemented and evaluated Logistic Regression, Random Forest, and Gradient Boosting models. Performed hyperparameter tuning for Gradient Boosting.
4. Model Improvement Addressed class imbalance using SMOTE (Synthetic Minority Over-sampling Technique). Implemented a Stacking Classifier combining Random Forest and Gradient Boosting models with a Logistic Regression meta-model. Conducted hyperparameter tuning for the Stacking Classifier.
5. SHAP Analysis Used SHAP (SHapley Additive exPlanations) to analyze feature importance and interpret model predictions.
6. Results Logistic Regression: Accuracy: 0.79, Precision: 0.62, Recall: 0.52, F1-Score: 0.56 Random Forest: Accuracy: 0.78, Precision: 0.63, Recall: 0.47, F1-Score: 0.54 Gradient Boosting: Accuracy: 0.79, Precision: 0.64, Recall: 0.48, F1-Score: 0.55 Tuned Gradient Boosting: Accuracy: 0.79, Precision: 0.63, Recall: 0.50, F1-Score: 0.56 Tuned GB with SMOTE: Accuracy: 0.77, Precision: 0.56, Recall: 0.65, F1-Score: 0.60 Stacking Classifier: Accuracy: 0.79, Precision: 0.64, Recall: 0.47, F1-Score: 0.54 Installation Clone the repository:

bash Copy code git clone https://github.com/your-username/telco-customer-churn-prediction.git cd telco-customer-churn-prediction Install required packages:

bash Copy code pip install -r requirements.txt Download the dataset from Kaggle and place it in the project directory:

Telco Customer Churn Dataset Usage Run the Jupyter notebook or Python scripts to replicate the analysis and results. Ensure you have the dataset in the correct directory.

Future Work Further feature engineering and exploration of additional models like XGBoost or LightGBM. More sophisticated hyperparameter tuning techniques. Investigating additional techniques for handling class imbalance.