📚 Semantic Book Recommender

Discover books that fit your mood, category, or desired tone with Semantic Book Recommender. This project combines the power of AI embeddings from Hugging Face, natural language processing, and a user-friendly Gradio interface to deliver a personalized book recommendation experience.

---

🚀 Features

• Natural Language Search: Find book recommendations based on a description or a query in natural language.
• Category Filtering: Refine your results by selecting specific book categories (e.g., Fiction, Self-help).
• Emotional Filtering: Get book recommendations based on emotional tones like Happy, Sad, Suspenseful, and more.
• User-friendly Interface: Built using Gradio, making it interactive and accessible for all users.

---

🛠️ How It Works

The Semantic Book Recommender uses the following steps:

1. Loads a dataset of books with metadata, emotional scores, and descriptions.
2. Leverages Hugging Face's sentence-transformers/all-MiniLM-L6-v2 model to generate embeddings for book descriptions.
3. Embeds user queries using the same model.
4. Retrieves the most semantically similar books using Chroma as the vector database engine.
5. Filters recommendations further based on user-selected category and emotional tone.

---

🖼️ Preview

---

📑 Project Structure

.
├── app.py                  # Main application file with Gradio logic
├── books_with_emotions.csv # Dataset containing books metadata, emotions, and thumbnails
├── tagged_description.txt  # File with tagged descriptions for semantic embeddings
├── requirements.txt        # List of Python dependencies
├── .env                    # Environment variables (for any OpenAI & Huggingface API keys)
└── README.md               # This README file

---

⚙️ Installation and Usage

To run the Semantic Book Recommender locally:

1. Clone the Repository:

   git clone https://github.com/asshejan/Book-Recommender.git
   cd Book-Recommender

2. Install Dependencies:

   pip install -r requirements.txt

3. Run the Application:

   python app.py

4. Open your browser and go to http://127.0.0.1:7860.

---

🛠️ Dependencies

Make sure you have the following libraries installed (or listed in requirements.txt):

• gradio
• langchain
• sentence-transformers
• pandas
• numpy
• python-dotenv
• chromadb

You can install them with:

pip install -r requirements.txt

---

🙌 Contribute

Feel free to contribute to the Semantic Book Recommender by:

• Submitting bug reports or feature requests.
• Adding new functionality (e.g., more emotional tones, better search algorithms).
• Cleaning and improving the existing code.

---

📧 Support

If you have any questions or run into issues, feel free to contact me:

• GitHub: Abu Sayeam Shejan

---

🌟 Acknowledgements

Special thanks to:

• Hugging Face for their awesome sentence-transformers.
• Gradio for making interactive machine learning applications easy to build.
• LangChain for simplifying AI workflows.

---

Enjoy discovering a world of books with Semantic Book Recommender! 📚✨