Overview:
BookScout is a full-stack, AI-driven book recommendation platform that helps users discover books they’ll love. By combining semantic similarity, genre filtering, and automated metadata enrichment, it delivers accurate and personalized recommendations.
Search & Recommend: Query by book title; TF-IDF + cosine similarity on descriptions and genres; optional genre filter checkboxes.
Metadata Enrichment: Uses Google Books API for metadata, Wikipedia fallback for plot summaries/categories, and zero-shot genre classification with facebook/bart-large-mnli.
Bulk Addition: Scrapes Goodreads lists for new titles, cleans/dedupes them, and adds in bulk with automated preprocessing.
Data Utilities: Scripts for scraping, cleaning, genre counting, and project structure visualization.
Backend: Flask, Pandas, BeautifulSoup, Hugging Face Transformers
Frontend: Chart.js, dark-mode UI
Data Storage: CSV-backed cache
BookScout demonstrates how AI and natural language processing can be applied to personalized recommendation systems, combining full-stack development with scalable data workflows.Â