Churn Prediction Analysis and Model Development

This project involves a comprehensive analysis of customer churn data and the development of a predictive model to identify customers who are at risk of churning. The project utilizes various data analysis techniques, statistical tests, and machine learning models.

Table of Contents

1. Project Overview
2. Dataset
3. Exploratory Data Analysis (EDA)
   • Customer Demographics
   • Churn Analysis
   • Product Usage
   • Financial Analysis
4. Predictive Modeling
   • Feature Engineering
   • Model Development
   • Hyperparameter Tuning
   • Model Evaluation
5. Requirements
6. How to Run
7. Results
8. Contributing
9. License

Project Overview

The goal of this project is to understand the factors influencing customer churn and develop a predictive model that can help identify customers who are at risk of leaving. The project includes:

• Data exploration and visualization.
• Statistical analysis to determine relationships between variables.
• Development and tuning of a machine learning model using XGBoost.
• Saving the trained model for future predictions.

Dataset

The dataset used in this project contains customer information such as age, gender, geographical location, balance, credit score, and whether the customer has churned or not.

Exploratory Data Analysis (EDA)

Customer Demographics

1. Age Distribution: The distribution of customers across different age groups was visualized using a histogram.

2. Gender Distribution: The gender distribution of customers was analyzed using a pie chart.

Churn Analysis

1. Churn Percentage: The percentage of customers who have churned was calculated.

2. Reasons for Churn: A chi-square test was performed to determine if there is a significant relationship between categorical variables (e.g., Geography, Gender, NumOfProducts) and customer churn.

3. Churn Patterns: Churn patterns were identified by analyzing the distribution of churned customers across different segments (e.g., Geography, Gender, Age).

Financial Analysis

1. Average Account Balance: The average balance of customers was calculated.

2. Financial Characteristics: The financial characteristics of churned vs. non-churned customers were compared using histograms.

Predictive Modeling

Feature Engineering

• Geography and Gender Encoding: The Geography column was converted to boolean values using one-hot encoding, and the Gender column was mapped to binary values.

Model Development

• Initial XGBoost Model: An XGBoost model was trained to compute feature importances, identifying the top 5 most significant predictors of customer churn.

Hyperparameter Tuning

• Grid Search: Hyperparameter tuning was performed using GridSearchCV to find the best model configuration.

Model Evaluation

• Classification Report: The final model was evaluated using a classification report, providing metrics such as precision, recall, and F1-score.

Requirements

To run this project, you need the following Python libraries:

pandas
matplotlib
scipy
xgboost
scikit-learn
joblib

How to Run

1. Clone the Repository: Clone this repository to your local machine.
2. Install Dependencies: Install the required Python libraries.
3. Run the Script: Execute the Jupyter notebook or Python script to perform the analysis and train the model.
4. View Results: The trained model will be saved as model.pkl, and you can use it for future predictions.

Results

• The top 5 most significant predictors of customer churn were identified.
• A tuned XGBoost model was developed, achieving high accuracy in predicting at-risk customers.

Contributing

If you would like to contribute to this project, please fork the repository and submit a pull request.