LLM Semantic Book Recommender

• Data Preprocessing: Cleaned and explored book metadata and descriptions.

• Vector Search: Implemented semantic search using vector embeddings to match user queries with relevant books.

• Zero-Shot Classification: Used LLMs to classify books as fiction or non-fiction without training data.

• Sentiment & Emotion Analysis: Extracted emotional tone (e.g., suspenseful, joyful) from book descriptions to enhance recommendations.

• Web App Deployment: Created an interactive Gradio interface for users to explore and filter book recommendations.