E-Commerce Product Delivery Prediction

This project predicts whether a product from an e-commerce company will be delivered on time or not. It also analyzes various factors affecting delivery and studies customer behavior. The project includes machine learning models for prediction and a Power BI dashboard for visualization.

GitHub Repository

All project files and detailed documentation are available on the associated GitHub profile.
GitHub Repository

Project Structure

ecomm-delivery-project/
│
├── app/
│   ├── static/
│   │   ├── css/
│   │   │   ├── style.css
│   │   │   ├── mediaqueries.css
│   │   ├── js/
│   │       ├── script.js
│   ├── templates/
│   │   ├── layout.html
│   │   ├── project3.html
│   ├── app.py
│   ├── utils.py
│   ├── static/
│   │   ├── data/
│   │       ├── compare_df.csv
│
├── Dockerfile
├── requirements.txt
├── README.md
└── .gitignore

How to Run Locally

1. Using Flask (app.py)

1. Clone the repository:

   git clone https://github.com/Aftabby/ecomm-delivery-project.git
   cd ecomm-delivery-project

2. Create a virtual environment and activate it:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the Flask application:

   python app/app.py

5. Open your browser and navigate to http://127.0.0.1:5000.

---

2. Using Docker

1. Clone the repository:

   git clone https://github.com/Aftabby/ecomm-delivery-project.git
   cd ecomm-delivery-project

2. Build the Docker image:

   docker build -t ecomm-delivery-project .

3. Run the Docker container:

   docker run -p 5000:5000 ecomm-delivery-project

4. Open your browser and navigate to http://127.0.0.1:5000.

---

Features

• Predicts on-time delivery using machine learning models.
• Analyzes customer behavior and product attributes affecting delivery.
• Includes a Power BI dashboard for interactive visualizations.

Machine Learning Models Used

• Random Forest Classifier
• Decision Tree Classifier
• Logistic Regression
• K Nearest Neighbors Classifier

License

This project is licensed under the MIT License. See the LICENSE file for details.