Book-Rating-Prediction-using-Recommender-Systems-for-goodreads.com

We have created various models to predict prospective user ratings for unread books for goodreads.com in Poetry genre. We Used methods such as item-item CF, Matrix factorization with neural CF, and Neural Matrix factorization to build models. We have achieved MAE of 0.779 and MSE of 0.964 signifying our predictions to be within error margin of 1 for a 0-5 rating system. Model results can be used to provide recommendations to users based on highest prospective rating as well as the next item prediction recommendation.