Multi Agent Recommender System (MARS)

A Telegram Bot and API that helps users with namaz-related questions using a multi-agent recommender architecture.
Explore the docs »
📺 Watch Demo Video
🌐 View Product

Table of Contents

1. Project Goal and Description
2. Project Context Diagram
3. Built With
4. Getting Started
5. Usage Guide
6. Feature Roadmap
7. Development (on github pages)
8. Quality Assurance (on github pages)
9. Build & Deployment (on github pages)
10. Architecture
11. License
12. Contacts
13. Acknowledgments

🧭 Project Goal and Description

The goal of this project is to build a multi-agent recommendation system that assists users with Islamic prayer guidance through an interactive Telegram bot. The system is backed by a FastAPI server and uses MongoDB to store users and conversations.

Key Objectives:

• Provide helpful recommendations and answers about namaz in a conversational way.
• Maintain and analyze user interactions to improve personalized responses.

System Structure:

• A Telegram bot (using python-telegram-bot) that communicates with users.
• A FastAPI backend that handles data storage, business logic, and API routing.
• A MongoDB database accessed via the async motor driver.

(back to top)

🧩 Project Context Diagram

(back to top)

🛠️ Built With

(back to top)

🛣️ Feature Roadmap

• Bot can understand the context
• Bot can understand the app structure
• User can create multiple conversation
• User can ask everything about namaz
• User can send voice messages
• Endpoint for Namaz.app
• Provide the explanation how model work

(back to top)

📘 Usage Guide

🤖 Telegram Bot

You can use our bot by launching it using the provided link. No further action is required.

🌐 FastAPI Backend

To use the recommender system you can contact us to get API endpoints.

(back to top)

🚀 Getting Started

Railway Deployment instructions

The Railway service is used to deploy our project. To repeat the project deployment, you can follow these steps:

1. Register or log in to your Railway account
2. Click to button to deploy new project
   
3. And choose project to deploy from the Github
4. Add enviroment variables:

• 4.1 Set your bot token
• 4.2 Set webhook URL. You can find it in Settings -> Networking section

6. Deploy project and make adjustments according to the logs

📦 Manual deployment

1. Clone the repo:

   git clone https://github.com/swp-team-1/mars_bot_1.5.git
   cd mars_bot_1.5

2. Create and activate a virtual environment:

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install the requirements:

   pip install -r requirements.txt

4. Create an .env file (copy from .env.example) and configure:

   BOT_TOKEN=your_telegram_bot_token
   MONGODB_URI=your_mongodb_connection_uri

5. Run services:

   uvicorn main:app --reload      # FastAPI backend
   python bot_main.py             # Telegram bot

(back to top)

🚀 Contributing

Contributions make the open source community an incredible place to learn and grow. Any help is greatly appreciated!

Have an idea to improve mars_bot_1.5? Fork the repo, make your changes, and open a pull request. Or just open an issue with the tag enhancement.

1. Fork the project
2. Create your branch (git checkout -b feature/MyFeature)
3. Commit your changes (git commit -m 'Add MyFeature')
4. Push (git push origin feature/MyFeature)
5. Open a pull request

(back to top)

🌟 Top Contributors

📄 License

Distributed under the MIT License. See LICENSE for more information.

(back to top)

📬 Contacts

• Black-persik – @black_persik
• tabletka_812 – @tabletka_812
• eerterr – @eerterr
• Leo_Vesin – @Leo_Vesin
• Desgun4ik – @Desgun4ik
• Project: github.com/swp-team-1/mars_bot_1.5

(back to top)

🙏 Acknowledgments

• python-telegram-bot – Telegram Bot Framework
• FastAPI – Web framework
• Motor – Async MongoDB driver
• python-dotenv – Manage environment variables
• contrib.rocks – Contributor avatars

(back to top)

📚 Documentation

Documenttation on Github Pages

• CONTRIBUTING.md
• Quality Attribute Scenarios
• Quality assurance
• CI Workflow
• CD Process

(back to top)

🏗 Architecture

• Static View
• Dynamic View
• Deployment View

(back to top)