Semantic Book Recommender with LLMs

This project develops a semantic book recommender system leveraging the power of Large Language Models (LLMs). It encompasses several key components:

Project Components

• Text Data Cleaning: Preparing raw text data for effective use in language models.
• Semantic (Vector) Search & Vector Database: Building a system to find semantically similar books using vector embeddings and storing them in a dedicated database.
• Text Classification with Zero-Shot Learning: Performing text categorization without prior training examples for specific categories, utilizing LLMs' inherent understanding.
• Sentiment Analysis & Emotion Extraction: Analyzing the emotional tone of text and extracting specific emotions using LLMs.
• Web Application with Gradio: Creating an interactive user interface using Gradio for users to easily get book recommendations.

---

Setup and Dependencies

This project was developed using Python 3.11. To run the project successfully, please ensure you have the following dependencies installed:

kagglehub pandas matplotlib seaborn python-dotenv langchain-community langchain-opencv langchain-chroma transformers gradio notebook ipywidgets

A requirements.txt file, containing all necessary project dependencies, is provided in this repository for easy installation. You can install them using pip:

pip install -r requirements.txt