Neural Movie Recommendation System

This is a movie recommendation system based on a Neural Collaborative Filtering model, served via a RESTful API using the Flask framework.

✨ Features

• Model Training: Trigger the model training process via an API call.
• Real-time Recommendations: Get instant movie recommendations for a specific user.
• RESTful API: Recommendation functionality is exposed through an easy-to-integrate API.
• Automated Documentation: Interactive API documentation is auto-generated using Flasgger (Swagger).

🛠️ Tech Stack

• Backend: Flask, Flask-CORS
• Machine Learning: PyTorch, Pandas, Scikit-learn
• API Docs: Flasgger
• Dependency Management: Poetry

🚀 Getting Started

1. Prerequisites

• Ensure you have Python 3.11+ installed.
• Ensure you have Poetry installed.

2. Installation

# Clone the repository
git clone https://github.com/YOUR_USERNAME/NeuralRec-System.git
cd NeuralRec-System

# Install dependencies using Poetry
poetry install

3. Download the Dataset

This project uses the MovieLens (small) dataset.

# Download and extract the dataset
wget https://files.grouplens.org/datasets/movielens/ml-latest-small.zip
unzip ml-latest-small.zip -d ml-latest-small/

# (Optional) Remove the zip file
rm ml-latest-small.zip

💡 Usage

1. Start the Service

poetry run python app.py

The service will be running at http://127.0.0.1:5001.

2. Browse API Documentation

After starting the service, you can view detailed information about all APIs and test them interactively at:

http://127.0.0.1:5001/docs/

📡 API Endpoints

POST /train

Triggers the model training process. The latest model is automatically saved upon completion.

• Request Body: (empty)
• Success Response:

  {
    "status": "success",
    "message": "Model training completed and saved"
  }

POST /recommend

Gets a list of movie recommendations for a given user ID.

• Request Body:

  {
    "userId": 1,
    "movieId": 2
  }

• Success Response:

  {
    "status": "success",
    "data": [10, 25, 150, 300, 550]
  }

POST /save

Manually saves the current state of the model.

• Request Body: (empty)
• Success Response:

  {
    "status": "success",
    "message": "Model saved successfully"
  }

📄 License

This project is licensed under the MIT License.