Pluto Plato

📚 Overview

Pluto plato Information Integration Application (IIA) course project: end-to-end ETL pipelines, schema matching, federated SQL querying, and AI-driven analytics for restaurant/vendor data. Built with Python, React, and LLM-powered natural-language interfaces.
The project focuses on integrating and analyzing information from various sources, particularly in the context of restaurants, dishes, and vendors.
It aims to create a system that can process and unify data from different formats and sources, making it accessible and useful for:

• Restaurant recommendations
• Menu analysis
• Vendor management

The project has the following robust objectives:

---

🧠 Objectives

• (a) Project Statement:
  Motivation: To create a scalable and dynamic system for real-world applications.
  Importance: Enhances data accessibility and utility for stakeholders such as restaurant owners, vendors, and end-users.
  Target Users: Restaurants, food delivery apps, vendors, and foodies.
  Focus: Innovative applications like real-time menu analytics and vendor comparison systems.

• (b) Underlying Data Sources:

  • Type: Relational databases (MySQL)
  • Schemas/APIs: Well-defined schemas for structured data exchange
  • Data Population: Populated via extraction scripts and document analysis
  • Communication Protocols: Standard SQL and REST APIs

• (c) Defining the Data Integration System:

  • Approach: Hybrid — combines structured integration with AI-based enhancements.

• (d) Schema Matching and Mapping:

  • Smart mapping between a global schema (with Restaurant and Dishes tables) and different vendor-specific schemas.

• (e) Query Interface, Query Federation, Execution, and Result Integration:

  • Allowing federated query execution across databases with result integration.

• (f) Query Analyzer and Decomposer:

  • Breaks down complex queries for optimized and efficient execution.

• (g) Query Federation with Load and Traffic Balancer:

  • Efficient load distribution across databases to minimize latency.

• (h) Aggregating and Displaying Results:

  • Merging, ranking, and presenting results seamlessly on the frontend.

---

🧩 Backend Functionality

The backend, primarily written in Python, handles core data processing and integration tasks.

Major Components:

• Data Extraction:
  addpdf_llm.py, extract_dish_info.py — extract restaurant and dish information from semi-structured sources like PDFs in context of a given sql query.

• Entity Analysis:
  entityanalysis.py, docanalysis.py — identify entities such as dish names, restaurant names, and vendors in context of a fiven sql query.

• Natural Language Processing:
  naturallanguageforadmin.py, naturalquery.py — process natural language queries from users/admins.

• Data Mapping and Integration:
  algomapping.py, contextfromsql.py — unify data by mapping different schemas, support addition and deletion.

• API Provisioning:
  app.py — defines and exposes backend routes for frontend consumption.

• Caching:
  .json files — store intermediate cache and metadata for faster processing.

Important Note

• Define a global schema containing at least two tables:
  Restaurant and Dishes.
• Add database connection info (user, password) in a global connection config.
• Databases must be on the same network (can be on different machines and different schemas).

---

🎨 Frontend Functionality

Located in the pluto-plato-frontend/ directory, the frontend is built using React.js and offers:

• User Interface:
  Displays restaurant, dish, and vendor data cleanly and intuitively.

• Search and Query Support:
  Users can search restaurants or dishes using natural language.

• Interactive Elements:
  Filter, sort, and navigate easily using dynamic UI components.

---

🚀 Getting Started

Prerequisites

Ensure you have installed:

• Python 3.8+
• Node.js 14+

---

🔧 Backend Setup

1. Navigate to the backend directory:

   cd backend

2. (Optional) Create and activate a virtual environment:

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

3. Install Python dependencies:

   pip install -r requirements.txt

4. Run the backend server:

   python app.py  # (or the appropriate entry point)

---

🎯 Frontend Setup

1. Navigate to the frontend directory:

   cd pluto-plato-frontend

2. Install Node.js dependencies:

   npm install

3. Start the development server:

   npm start

---

🌐 Accessing the Application

Open your browser and visit:

http://localhost:3000

(or the port specified in your configuration)

---

✨ Features

• 🔍 Extraction of restaurant and dish information from semi-structured sources
• 🗣️ Natural language query support
• 🔗 Data integration across multiple vendors
• 🖥️ User-friendly frontend interface
• ⚡ Fast caching and smart query decomposition
• 🔄 Federated query processing and load balancing

---

🤝 Credits

This repository contains code created through the combined efforts of:

• @aditya22041
• @A-WASIF
• @aastha1708

Uploaded with their permission, acknowledging their original contributions. 🚀