title	AI Book Recommender
emoji	📚
colorFrom	purple
colorTo	blue
sdk	docker
pinned	false
license	mit

📚 AI Book Recommender

Discover your next favorite book with AI-powered recommendations based on semantic search and emotional tone analysis

🚀 Try the Live App

Features • Installation • Usage • Technologies • Project Structure

---

🌟 Features

• 🤖 AI-Powered Semantic Search - Uses OpenAI embeddings to understand the meaning behind your book preferences
• 🎭 Emotion-Based Filtering - Filter books by emotional tone (Happy, Sad, Angry, Suspenseful, Surprising)
• 📑 Category Filtering - Browse books by category for more targeted recommendations
• ⚡ Fast Vector Database - Persistent Chroma vector database for quick similarity searches
• 🎨 Beautiful UI - Modern, responsive Gradio interface with gradient themes
• 📊 Rich Book Metadata - Displays book covers, titles, authors, and descriptions

🎯 How It Works

1. Describe Your Ideal Book - Enter a natural language description of what you're looking for
2. Filter by Preferences - Optionally select a category and emotional tone
3. Get Personalized Recommendations - Receive 16 curated book suggestions with covers and descriptions

The system uses:

• Semantic Search: Converts book descriptions to vector embeddings for intelligent matching
• Emotion AI: Analyzes emotional content in book descriptions (joy, sadness, anger, fear, surprise)
• Hybrid Filtering: Combines semantic similarity with category and emotion filters

🚀 Installation

Prerequisites

• Python 3.11 or higher
• OpenAI API key (Get one here)

Setup

1. Clone the repository

   git clone https://github.com/yourusername/Book-recommender.git
   cd Book-recommender

2. Create a virtual environment (recommended)

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

3. Install dependencies

   pip install -r requirements.txt

4. Set up environment variables

   Create a .env file in the project root:

   OPENAI_API_KEY=your_api_key_here

5. Prepare the data

   The first time you run the application, it will automatically create a vector database from the book descriptions. This may take a few minutes.

💻 Usage

Running the Application

python gradio_dashboard.py

The application will launch in your default web browser at http://127.0.0.1:7860

Example Queries

Try these example searches to get started:

• "A coming-of-age story about friendship and self-discovery" → Category: Fiction, Tone: Happy
• "A historical novel set during World War II" → Category: Nonfiction
• "An epic fantasy adventure with dragons and magic" → Category: Fiction, Tone: Surprising
• "A psychological thriller with unexpected twists" → Category: Fiction, Tone: Suspenseful

Data Exploration

Explore the data cleaning and analysis process in the Jupyter notebook:

jupyter notebook data-exploration.ipynb

🛠️ Technologies

Technology	Purpose
Python 3.11+	Core programming language
Gradio	Interactive web UI framework
LangChain	Framework for building LLM applications
OpenAI Embeddings	Text vectorization for semantic search
Chroma	Vector database for efficient similarity search
Transformers (Hugging Face)	Emotion analysis on book descriptions
Pandas & NumPy	Data manipulation and analysis
Matplotlib & Seaborn	Data visualization

📁 Project Structure

Book-recommender/
│
├── gradio_dashboard.py          # Main Gradio application
├── data-exploration.ipynb       # Data cleaning & EDA notebook
├── requirements.txt             # Python dependencies
│
├── books_final.csv              # Cleaned book dataset
├── tagged_descriptions.text     # Book descriptions for embeddings
│
├── chroma_db/                   # Persistent vector database (auto-generated)
└── .env                         # Environment variables (not tracked)

📊 Dataset

The project uses a curated dataset of books containing:

• ISBN-13 - Unique book identifier
• Title - Book title
• Authors - Author(s) name(s)
• Description - Book synopsis/summary
• Categories - Book genre/category
• Thumbnails - Cover image URLs
• Emotion Scores - Joy, sadness, anger, fear, surprise metrics

🎨 Features in Detail

Semantic Search

The system converts your natural language query into a vector embedding and finds books with similar meaning in their descriptions, even if they don't share the exact same words.

Emotion Analysis

Each book description is analyzed for emotional content using transformer models, allowing you to find books that match your desired mood.

Persistent Vector Database

The Chroma database saves embeddings locally, so subsequent runs are much faster—no need to re-embed thousands of books each time!

🤝 Contributing

Contributions are welcome! Here are some ways you can help:

• 🐛 Report bugs and issues
• 💡 Suggest new features
• 📝 Improve documentation
• 🔧 Submit pull requests

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Book data sourced from Kaggle
• OpenAI for embeddings API
• Gradio team for the amazing UI framework
• Hugging Face for transformer models

---

Built with ❤️ by Zoro-chi

⭐ Star this repo if you found it helpful!