📚 AI Powered Book Recommender

An intelligent book recommendation system powered by AI and NLP techniques. This project combines data cleaning, semantic search, classification, sentiment analysis, and a simple web app to help users discover books based on meaning, mood, and categories rather than just keywords.

🚀 Features

✨ Text Data Cleaning

Preprocessed raw text data for better downstream tasks.

Code: data-exploration.ipynb

Semantic (Vector) Search

Built a vector database to find books similar to a natural language query.

Example: "a book about a person seeking revenge" → returns similar recommendations.

Code: vector-search.ipynb

Text Classification (Fiction / Non-Fiction)

Classified books using zero-shot classification with LLMs.

Enables users to filter books by type.

Code: text-classification.ipynb

Sentiment & Emotion Analysis

Extracted emotions and tone (suspenseful, joyful, sad, etc.) from book descriptions.

Allows users to sort by mood.

Code: sentiment-analysis.ipynb

Web Application with Gradio

Built an interactive web app for users to search & get recommendations.

Code: gradio-dashboard.py

TECH STACK

Programming Language: Python

Libraries & Frameworks:

Pandas

Matplotlib

Seaborn

python-dotenv

Transformers

Gradio

LangChain Community

Chroma

Jupyter Notebook

ipywidgets

kagglehub

📊 Data

Book datasets are downloaded from Kaggle.

Processed datasets (cleaned & labeled versions) are included in this repo for quick use.