Customer Churn Prediction Project

Project Overview

This project aims to analyse and predict customer churn for a telecom company using the Telco Customer Churn dataset. The primary goal is to identify the factors that contribute to customer churn and develop a predictive model to forecast which customers are likely to leave the company. The insights gained from this analysis can help the company implement targeted retention strategies.

Project Structure

customer_churn_analysis/
│
├── data/
│ ├── raw/
│ │ └── Telco_customer_churn.xlsx
│ ├── processed/
│ │ ├── processed_data.csv
│ │ └── enhanced_data.csv
│
├── src/
│ ├── data_preprocessing.py
│ ├── exploratory_data_analysis.py
│ ├── feature_engineering.py
│ ├── model.py
│ ├── evaluate.py
│ └── gen_report.py
│
├── reports/
│ ├── figures/
│ │ ├── confusion_matrix.png
│ │ ├── roc_curve.png
│ │ ├── lots of other graphs
│ └── generated_report.pdf
│ └── classification_report.txt
│ └── summary_statistics.csv
│
├── logs/
│ ├── project.log
│ └── run_evaluate_report.log
│
├── README.md
├── requirements.txt
├── run_all.py
└── run_evaluate_report.py

Scripts and Their Functions

1. data_preprocessing.py:

   • Loads the raw data, preprocesses it by handling missing values and encoding categorical variables, and saves the processed data.

2. exploratory_data_analysis.py:

   • Performs exploratory data analysis (EDA) on the processed data, generates visualizations, and saves them to the reports directory.

3. feature_engineering.py:

   • Engineers new features from the processed data and saves the enhanced dataset.

4. model.py:

   • Splits the enhanced data into training and testing sets, handles class imbalance, trains a logistic regression model, evaluates the model, and saves the trained model.

5. evaluate.py:

   • Loads the trained model and test data, evaluates the model performance, generates relevant metrics and visualizations, and saves them to the reports directory.

6. gen_report.py:

   • Compiles the results of the model evaluation into a comprehensive PDF report.

7. run_all.py:

   • Runs all the scripts in the correct sequence: data preprocessing, feature engineering, model building, model evaluation, and report generation.

8. run_evaluate_report.py:

   • Runs only the model evaluation and report generation scripts.

How to Run the Scripts

1. Install Dependencies:

   • Ensure you have Python installed. You can create a virtual environment and install the required dependencies using the following commands:

   python -m venv venv
   source venv/bin/activate # On Windows, use `venv\Scripts\activate`
   pip install -r requirements.txt

2. Run All Scripts:

   • To run the entire project from data preprocessing to report generation, execute:

   python run_all.py

3. Run Model Evaluation and Report Generation Only:

   • To run only the model evaluation and report generation scripts, execute:

   python run_evaluate_report.py

Dependencies

• Python 3.x
• pandas
• numpy
• matplotlib
• seaborn
• scikit-learn
• imbalanced-learn
• fpdf
• joblib

You can install these dependencies using the provided requirements.txt file.

Data Description

The dataset contains 7043 observations with 33 variables (features), including customer demographic information, services subscribed to, account information, and churn details. Key features include:

• Customer Demographics: Gender, Senior Citizen, Partner, Dependents
• Service Information: Phone Service, Multiple Lines, Internet Service, Online Security, Online Backup, Device Protection, Tech Support, Streaming TV, Streaming Movies
• Account Information: Contract, Paperless Billing, Payment Method, Monthly Charges, Total Charges, Tenure Months
• Churn Details: Churn Label, Churn Value, Churn Score, CLTV, Churn Reason

Objectives

1. Data Preprocessing: Clean and preprocess the data, handle missing values, and encode categorical variables.
2. Exploratory Data Analysis (EDA): Understand the data distribution and identify patterns and correlations.
3. Feature Engineering: Create new features to improve model performance.
4. Model Building: Develop a predictive model to forecast customer churn using various machine learning algorithms.
5. Model Evaluation: Assess the model's performance using appropriate metrics and validate its predictive power.
6. Reporting and Visualization: Generate insightful visualizations and a comprehensive report to communicate findings and recommendations.

Documentation and Logging

• Each script logs its progress and any issues encountered in the logs directory.
• The final report and visualizations are saved in the reports directory.