AI-Powered Search & Aggregation Tool

🌟 Overview

This is a full-stack AI-powered search and aggregation tool that takes user queries, retrieves results from Google and YouTube, and presents them in a structured React UI. The backend is built with FastAPI (Python), and the frontend is developed using React.js.

✅ Features:

• AI-Powered Search Expansion: Enhances user queries using NLP models.
• Google & YouTube Integration: Fetches search results from multiple sources.
• Multithreading Optimization: Improves speed using parallel processing.
• Sorting & Filtering: Allows users to customize search results.
• Dark Mode UI: Aesthetic, user-friendly design similar to Google Search.
• Toggle AI Search: Allows users to switch between AI-enhanced search and direct search.

---

🛠️ Tech Stack

Technology	Description
Frontend	React.js, CSS
Backend	FastAPI (Python)
Database	Not required (API-based search)
APIs Used	Google Search (SerpAPI), YouTube Data API
AI Model	Hugging Face Transformers (T5, BART)
Deployment	Vercel (Frontend), Render (Backend)

---

🚀 Setup & Installation

Clone the Repository

git clone https://github.com/G26karthik/Ai-search-tool.git
cd your-repository-name

🖥️ Backend Setup (FastAPI)

📌 Navigate to the backend folder:

cd backend

📌 Install dependencies:

pip install -r requirements.txt

📌 Set up environment variables: Create a .env file inside backend/ and add:

SERPAPI_KEY=your_serpapi_key
YOUTUBE_API_KEY=your_youtube_api_key

Run the FastAPI server:

uvicorn main:app --reload

📌 Test the API in browser:

http://127.0.0.1:8000/docs

🌐 Frontend Setup (React.js)

📌 Navigate to the frontend folder:

cd search-tool

📌 Install dependencies:

npm install

📌 Set up environment variables: Create a .env file inside search-tool/ and add:

REACT_APP_BACKEND_URL=http://127.0.0.1:8000

📌 Start the frontend:

npm start

📌 Open in browser:

http://localhost:3000

📊 Features & Functionality

1️⃣ AI Query Expansion

• Uses NLP models (T5 or BART) from Hugging Face to expand user queries.
• Helps generate more relevant search terms to improve results.
• Can be toggled ON/OFF using a toggle button:
  • AI Mode ON: Query is expanded before searching.
  • AI Mode OFF: Uses the exact user query without modification.
• Alert message appears when AI search is enabled, informing users that accuracy may vary.

---

2️⃣ Google & YouTube Search

• Google Search:
  • Uses SerpAPI to fetch organic search results.
  • Returns title, snippet, and link for each result.
• YouTube Search:
  • Uses YouTube Data API v3 to fetch relevant videos.
  • Displays video title, description, and link to watch.

---

3️⃣ Sorting & Filtering

• Users can sort results based on:
  • 🔍 Relevance (Default)
  • 📅 Date (Latest results first)
  • 🌐 Source (Group Google & YouTube separately)
• Filters help refine search results for better accuracy.

---

4️⃣ UI/UX Similar to Google

• Search Bar in the Center:
  • Neatly styled, like Google Search.
  • Toggle button below it to switch AI search ON/OFF.
• Results Displayed in Two Columns:
  • Left Side → 🌍 Google Search Results.
  • Right Side → 🎥 YouTube Search Results.
• Clean & Responsive Design:
  • Fully mobile-friendly.
  • Dark mode UI for better readability.

---

5️⃣ Multithreaded Backend for Faster Searches

• Uses Python’s FastAPI + Multithreading to speed up search requests.
• Google & YouTube searches run in parallel, reducing load time.

---

Video Demo

demo.Video.mp4

---

📑 Challenges & Future Improvements

✅ Challenges Faced:

• Query expansion sometimes changes the intent of the search.
• API rate limits for SerpAPI & YouTube Data API.
• Deployment issues with backend on Render, so it's running locally.

🚀 Future Improvements:

• Expand Search Sources:

  • Integrate additional platforms such as LinkedIn, Reddit, and Twitter for a more comprehensive search experience.

• Enhanced Ranking & Filtering:

  • Develop AI-based ranking algorithms to better prioritize the most relevant results.
  • Implement customizable filtering options based on user preferences.

• Database Integration:

  • Cache frequently searched queries and results to improve response times.
  • Store user search history and preferences for personalized recommendations.

• User Authentication:

  • Implement user accounts and secure login systems to provide personalized search experiences and save search histories.

• Performance Optimization:

  • Further optimize API calls and explore serverless architectures to scale with demand.

• UI/UX Enhancements:

  • Improve responsiveness and design for mobile devices.
  • Add interactive elements and animations for a smoother user experience.

🤝 Contributing

To contribute to this project, please follow these steps:

1️⃣ Fork the repository on GitHub.

cd your-repository-name

2️⃣ Create a new branch for your feature or bugfix.

git checkout -b feature-branch

3️⃣ Make your changes and commit them.

git add .
git commit -m "Description of your changes"

4️⃣ Push your branch to GitHub.

git push origin feature-branch

5️⃣ Open a Pull Request (PR) for review.

📜 License

This project is licensed under the MIT License. You are free to use, modify, and distribute this project with proper attribution.