Semantic Book Recommender

AI-driven application that recommends books based on emotions detected in user input and whether the user prefers fiction or non-fiction. This project leverages both OpenAI and Hugging Face APIs to perform text classification and zero-shot classification, respectively.

Table of Contents

• Overview
• Features
• Installation
• Usage
• Project Structure
• Technical Details
• Contact

---

Overview

The Semantic Book Recommender uses Natural Language Processing (NLP) techniques to analyze user input and suggest books. It relies on:

1. Emotion Detection: Determines the emotional tone (e.g., joy, sadness) of the user’s input.
2. Genre Classification: Uses zero-shot classification to categorize the user's query into "Fiction" or "Nonfiction."

By combining these insights, the application surfaces a curated list of book recommendations.

---

Features

• Emotion Analysis: Powered by j-hartmann/emotion-english-distilroberta-base for robust emotion classification.
• Zero-Shot Classification: Uses facebook/bart-large-mnli to distinguish between Fiction and Nonfiction.
• Interactive Dashboard: Built with Gradio for a user-friendly interface.
• CSV-based Book Data: Pre-processed and stored in books_with_emotions.csv and books_with_categories.csv.

---

Installation

1. Clone the Repository

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

2. Create a Virtual Environment (Optional but Recommended)

   python3 -m venv venv
   source venv/bin/activate  # On macOS/Linux
   # or
   venv\Scripts\activate     # On Windows

3. Install Dependencies

   pip install -r requirements.txt

4. Set Up API Keys
   • OpenAI: Create a .env file or export your API key as an environment variable:

   OPENAI_API_KEY="YOUR_OPENAI_API_KEY"

   • Hugging Face: If required, configure your token similarly:

   HUGGINGFACEHUB_API_TOKEN="YOUR_HF_API_KEY"
   

---

Usage

1. Run the Gradio Dashboard

   python3 gradio-dashboard.py

2. Access the Application
   • Copy the URL printed in the console (something like http://127.0.0.1:7860).
   • Open it in your browser.
3. Interact with the Dashboard
   • Provide a short description of a book you’re looking for, or a phrase that captures your mood.
   • Choose optional categories or let the model handle them automatically.
   • View the recommendations and explore the listed books.

---

Project Structure

book_recommender/
├── books_cleaned.csv
├── books_with_categories.csv
├── books_with_emotions.csv
├── cover-not-found.jpg
├── data-exploration.ipynb
├── gradio-dashboard.py
├── requirements.txt
├── sentiment-analysis.ipynb
├── tagged_description.txt
├── text-classification.ipynb
├── vector-search.ipynb
└── ...

• gradio-dashboard.py: Main file to launch the Gradio interface.
• books_with_emotions.csv and books_with_categories.csv: Data files used for recommendations.
• .ipynb notebook: Exploratory notebooks showcasing data cleaning, text classification, sentiment analysis, and vector search.

---

Technical Details

1. Emotion Classification

   from transformers import pipeline
   classifier = pipeline("text-classification", 
                      model="j-hartmann/emotion-english-distilroberta-base")
   # Analyzes the user's text and infers the dominant emotion.

2. Zero-Shot Classification

   from transformers import pipeline
   
   fiction_categories = ["Fiction", "Nonfiction"]
   pipe = pipeline("zero-shot-classification",
                model="facebook/bart-large-mnli")
   # Classifies user input to decide whether they want Fiction or Nonfiction.

3. Integration with OpenAI
   • OpenAI is used for any language model queries or text embeddings (depending on your configuration with langchain).
4. Gradio Frontend
   • Provides a web interface for user input and displays the recommended books.
5. Data
   • The CSV files contain pre-labeled or pre-processed metadata about books (title, author, categories, etc.).

---

Contact

• Email: kennygarcia15@yahoo.com
• LinkedIn: linkedin.com/in/kennygarcia15 Feel free to reach out if you have questions, suggestions, or feedback!

---

Thank you for checking out the Semantic Book Recommender. Enjoy exploring new reads tailored to your mood and preferences!