https://github.com/sayande01/customer_churn_dnn

A Python project to predict customer churn using a neural network. Includes data preprocessing, exploratory analysis, and a TensorFlow-based model for binary classification. Ideal for understanding churn behavior and building predictive solutions.
https://github.com/sayande01/customer_churn_dnn

deep-learning neural-network

Last synced: 2 months ago
JSON representation

A Python project to predict customer churn using a neural network. Includes data preprocessing, exploratory analysis, and a TensorFlow-based model for binary classification. Ideal for understanding churn behavior and building predictive solutions.

• Host: GitHub
• URL: https://github.com/sayande01/customer_churn_dnn
• Owner: sayande01
• Created: 2025-01-23T19:15:37.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2025-01-23T19:21:14.000Z (5 months ago)
• Last Synced: 2025-02-13T02:38:15.034Z (4 months ago)
• Topics: deep-learning, neural-network
• Language: Jupyter Notebook
• Homepage:
• Size: 524 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        

# Customer Churn Prediction Using Neural Networks

This repository contains a Python project for predicting customer churn using a neural network. It utilizes a dataset of customer information, cleans and preprocesses the data, and applies a TensorFlow-based neural network model to classify churn probability.

---

## Table of Contents

- [Project Overview](#project-overview)

- [Dataset](#dataset)

- [Technologies Used](#technologies-used)

- [Project Workflow](#project-workflow)

- [Model Architecture](#model-architecture)

- [Installation](#installation)

- [Usage](#usage)

- [Results](#results)

- [Contributing](#contributing)

- [License](#license)

---

## Project Overview

Customer churn is a critical metric for businesses, especially subscription-based services. This project aims to predict churn based on customer data using a machine-learning approach. 

The project workflow includes:

1. Exploratory Data Analysis (EDA) for data understanding.

2. Data preprocessing to clean and prepare the dataset.

3. Building and training a neural network to classify churn outcomes.

4. Evaluating model performance to ensure accuracy and reliability.

---

## Dataset

The dataset used for this project contains customer information such as:

- **Demographics**: Gender, SeniorCitizen, Partner, Dependents.

- **Services**: PhoneService, MultipleLines, InternetService, StreamingTV, etc.

- **Account Information**: Tenure, Contract Type, Payment Method, Monthly and Total Charges.

- **Churn**: Binary indicator of whether the customer has churned.

**Dataset Summary**:

- **Rows**: 7,043

- **Columns**: 21

- No missing or duplicate values were found.

---

## Technologies Used

- **Programming Language**: Python

- **Libraries and Frameworks**:

  - Data Processing: `pandas`, `numpy`

  - Visualization: `matplotlib`, `seaborn`

  - Machine Learning: `TensorFlow`, `Keras`

---

## Project Workflow

1. **Data Loading and Cleaning**:

   - Load the dataset.

   - Handle missing values and incorrect data types.

2. **Exploratory Data Analysis**:

   - Understand distributions and correlations.

   - Visualize features affecting churn.

3. **Feature Engineering**:

   - Normalize and encode categorical data.

   - Prepare the dataset for neural network input.

4. **Model Building**:

   - A 3-layer neural network with `relu` activation.

   - Binary cross-entropy as the loss function.

5. **Training and Evaluation**:

   - Train the model on the dataset for 100 epochs.

   - Evaluate model accuracy and loss.

---

## Model Architecture

The neural network architecture:

- **Input Layer**: 23 features.

- **Hidden Layers**: 

  - First layer: 23 neurons, ReLU activation.

  - Second layer: 15 neurons, ReLU activation.

- **Output Layer**: 1 neuron, Sigmoid activation.

- **Optimizer**: Adam with default learning rate.

- **Loss Function**: Binary Crossentropy.

---

## Usage

1. Prepare your dataset (`customer_churn.csv`).

2. Load the notebook or script.

3. Train the model using:

   ```python

   model.fit(X_train, y_train, epochs=100)

   ```

4. Evaluate the model and use it to predict churn probabilities.

---

## Results

The model achieves **[Add Accuracy Here]%** accuracy after training for 100 epochs. Further optimization can be done by tuning hyperparameters or applying additional feature engineering.

---

## Contributing

Contributions are welcome! Please fork the repository and submit a pull request for any enhancements or bug fixes.