README.md

metadata

title: Book Recommender
emoji: 📉
colorFrom: gray
colorTo: blue
sdk: streamlit
sdk_version: 1.41.1
app_file: app.py
pinned: false
short_description: A basic book recommender system

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

Content-Based Book Recommender System

This is a content-based book recommender system built using Streamlit. It takes a book title as input and generates up to 5 recommendations based on content similarity. The recommendations combine textual similarity (TF-IDF) and semantic similarity (Sentence Transformers). The system also handles typos and invalid inputs.

---

Features

1. Content-Based Recommendations:

   • Combines TF-IDF and semantic similarity to recommend books.
   • Handles typos and approximate matches using fuzzy matching.

2. Error Handling:

   • Ensures graceful failure for invalid inputs or system errors.
   • Provides fallback recommendations when similarity scores are insufficient.

3. Interactive UI:

   • Built with Streamlit for a user-friendly interface.

Dependencies and Functionality

1. pandas: Handles data loading, preprocessing, and manipulation.

2. scikit-learn: Provides TF-IDF vectorization and NearestNeighbors for similarity search.

3. sentence-transformers: Generates semantic embeddings for textual data.

4. numpy: Used for numerical computations like handling embeddings and arrays.

5. streamlit: Creates the interactive UI for the application.

6. rapidfuzz: Implements fuzzy matching for typos and approximate title matches.

7. re: Cleans special characters from book titles.