https://github.com/parthds02/customer-churn-prediction-with-streamlit
Customer churn prediction is a vital tool for businesses to identify customers likely to leave their service. In this project, I build a machine learning pipeline to predict customer churn, integrating data analysis, feature engineering, modeling, and Demonstrate as a Streamlit web application.
https://github.com/parthds02/customer-churn-prediction-with-streamlit
churn-analysis churn-prediction decision-trees eda knn logistic-regression machine-learning random-forest streamlit support-vector-classifier vscode
Last synced: 3 months ago
JSON representation
Customer churn prediction is a vital tool for businesses to identify customers likely to leave their service. In this project, I build a machine learning pipeline to predict customer churn, integrating data analysis, feature engineering, modeling, and Demonstrate as a Streamlit web application.
- Host: GitHub
- URL: https://github.com/parthds02/customer-churn-prediction-with-streamlit
- Owner: ParthDS02
- License: mit
- Created: 2024-12-15T09:52:32.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2024-12-15T10:16:00.000Z (6 months ago)
- Last Synced: 2025-02-08T20:12:50.404Z (5 months ago)
- Topics: churn-analysis, churn-prediction, decision-trees, eda, knn, logistic-regression, machine-learning, random-forest, streamlit, support-vector-classifier, vscode
- Language: Jupyter Notebook
- Homepage:
- Size: 104 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Customer Churn Prediction with Streamlit
## Overview
Customer churn prediction is a vital tool for businesses to identify customers likely to leave their service. In this project, I build a machine learning pipeline to predict customer churn, integrating data analysis, feature engineering, modeling, and Demonstrate as a Streamlit web application.
### Project Highlights:
- **Dataset Exploration**: Analyzed key patterns and trends in customer behavior.
- **Data Cleaning and Feature Engineering**: Prepared the data for accurate model training.
- **Model Training and Evaluation**: Tested five models, achieving up to **91.5% accuracy**.
- **Interactive Web Application**: Created a user-friendly app with Streamlit for real-time churn predictions.
## Tools and Technologies
- **Language**: Python (v3.11.5)
- **IDE**: Visual Studio Code
- **Libraries**:
- **Data Manipulation**: Pandas, NumPy
- **Visualization**: Matplotlib
- **Machine Learning**: Scikit-learn
- **App Development**: Streamlit
- **Model Persistence**: Joblib
## Project Workflow
### 1. Dataset Exploration
- Loaded and analyzed the dataset, focusing on the target variable (**churn**).
### 2. Data Cleaning
- Handled missing values and duplicates to ensure data integrity.
### 3. Data Visualization
- Created visualizations like pie charts and histograms to understand key trends.
### 4. Feature Engineering
- Encoded categorical variables for machine learning models.
### 5. Model Training and Evaluation
| Model | Accuracy (%) |
|--------------------------|--------------|
| Logistic Regression | **91.50** |
| K-Nearest Neighbors | 91.00 |
| Support Vector Classifier| 91.00 |
| Decision Tree | 85.00 |
| Random Forest Classifier | 89.00 |
### 6. Streamlit App Creation
- Built an interactive app for churn predictions.
### 7. Demostration
- Demostration the app locally.
## How to Run the Project
1. Clone the repository:
```bash
git clone https://github.com/your_username/churn-prediction.git
cd churn-prediction
```
2. Install dependencies:
```bash
pip install -r requirements.txt
```
3. Run the Streamlit app:
```bash
streamlit run app.py
```
4. Open the app in your browser to input customer data and receive predictions.
## Directory Structure
```
churn-prediction/
|-- data/ # Contains the dataset
|-- notebooks/ # Jupyter Notebooks for EDA and model training
|-- app.py # Streamlit application
|-- models/ # Saved ML models and scaler using Joblib
|-- README.md # Project documentation (this file)
```
## Screenshots
Include screenshots of:
- Data visualizations (pie chart, Bar chart, histogram)
- Streamlit app interface (input form, prediction result)
### Data Visualization Example
### App Interface
## Key Takeaways
- Logistic Regression performed the best with **91.5% accuracy**.
- The Streamlit app offers an intuitive way to leverage machine learning for business insights.
## License
This project is licensed under the MIT License. See the `LICENSE` file for details.
## Contributing
Feel free to fork the repository and submit a pull request to suggest improvements.
## Future Enhancements
- Deploy the app on a cloud platform like AWS or Azure.
- Incorporate additional features to improve model accuracy.