Chat_Assistant

🚀 Python-based Chat Assistant for SQLite Database
🔗 Live API: Chat Assistant API

📌 Overview

This project is a Flask-based Chat Assistant that interacts with an SQLite database to answer user queries. It accepts natural language input, converts it into SQL queries, fetches data, and responds in a clear, formatted manner.

✅ Features

✔ Accepts natural language queries.
✔ Converts user input into SQL queries.
✔ Retrieves relevant data from an SQLite database.
✔ Handles invalid queries & errors gracefully.
✔ Supports various queries, including:

• "Show me all employees in the [department] department."
• "Who is the manager of the [department] department?"
• "List all employees hired after [date]."
• "What is the total salary expense for the [department] department?"

🔧 Technologies Used

• Python (Core programming language)
• SQLite (Database management)
• Flask (Web API)
• NLTK (For natural language processing, if needed)
• SQLAlchemy (For database interaction)

🚀 Getting Started

1️⃣ Clone the Repository

git clone https://github.com/Subramanyabs15/Chat_Assistant.git
cd Chat_Assistant

2️⃣ Set Up a Virtual Environment

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

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Prepare the SQLite Database

Ensure employees.db exists with this schema:

CREATE TABLE Employees (
    ID INTEGER PRIMARY KEY,
    Name TEXT NOT NULL,
    Department TEXT NOT NULL,
    Salary INTEGER NOT NULL,
    Hire_Date TEXT NOT NULL
);

CREATE TABLE Departments (
    ID INTEGER PRIMARY KEY,
    Name TEXT NOT NULL,
    Manager TEXT NOT NULL
);

If missing, create it manually or use create_db.py.

5️⃣ Run the Assistant

python app.py

The API will be accessible at:

http://127.0.0.1:5000

If using Railway, use your live URL:

https://chatassistant-production-2fb6.up.railway.app

🌐 API Endpoints

Method	Endpoint	Description
GET	/	Returns "Chat Assistant API is running!"
POST	/chat	Accepts a user query and returns results

📌 Example API Requests

📝 1️⃣ Using curl (Command Line)

curl -X POST "https://chatassistant-production-2fb6.up.railway.app/chat" \
-H "Content-Type: application/json" \
-d '{"query": "Who is the manager of the Sales department?"}'

📝 2️⃣ Using Postman

Method: POST
URL: https://chatassistant-production-2fb6.up.railway.app/chat
Headers: { "Content-Type": "application/json" }
Body (JSON):

{
    "query": "Who is the manager of the Sales department?"
}

✅ Expected Response:

{"response": [["Alice"]]}

🔍 How It Works

1. User enters a query in natural language.
2. Query gets converted into SQL using query_parser.py.
3. SQLite database fetches relevant data.
4. The response is formatted & returned in JSON.

⚠️ Error Handling

✔ Handles invalid department names (returns appropriate message).
✔ Gracefully manages incorrect query formats.
✔ Returns “No results found” for queries with no matching data.

🚀 Known Limitations & Future Improvements

🚧 Limited NLP Support: Complex sentence handling can be improved.
🚧 Enhancements Needed for Multi-Condition Queries.
🚧 User Authentication: Not yet implemented, could be a future feature.

🤝 Contribution

Feel free to fork the repository and submit pull requests with improvements! Your contributions are welcome! 😊🎯

License

This project is open-source and available under the MIT License.