A book recommendation system using item-item collaborative filtering. It is a simple implemenation of collaborative filtering approach for recommendation systems. The system recommends books to the user based on their recent read by using k-Nearest neighbors approach on user-item sparse matrix.

The dataset used is goodreads-10k dataset that contains 6 million ratings for ten thousand most popular books (https://github.com/zygmuntz/goodbooks-10k).

• python >= 3.6

• node.js >= 10.12.0

pip install --user virtualenv
virtualenv env
source env/bin/activate

git clone https://github.com/prati1/suggest-a-book.git
cd suggest-a-book/frontend
npm install
cd ../recommender-api
pip install -r requirements.txt

Go to recommender-api folder

python sparse_matrix_generator.py

python recommender_api.py

Go to frontend folder

npm start