Utilize Deep learning to increase the performance of a Book recommender system using the Item-based Collaborative Filtering

Item-based Collaborative Filtering is a common and efficient approach for recommendation problems. In this study, we have investigated the power of deep learning in textual feature extraction and applied this advantage to a high-performance item-based collaborative filtering recommender system. The proposed approach has been experienced on book datasets added by texts collected from famous book review sites. The experiment proves that the proposed model has better performance thanks to the contribution of the new item profile process method based on Deep Learning.