DCCR: A Deep Collaborative Conjunctive Recommender System for Accurate Rating Predictions

The Deep Collaborative Conjunctive Recommender (DCCR) model is presented in this research with the objective for enhancing the accuracy of rating prediction in recommendation systems. DCCR generalizes conjunctive preferences using matrix factorization techniques and captures intricate user-item interactions by fusing deep learning with collaborative filtering. A conjunctive layer synthesizes multi-attribute suggestions for accurate rating predictions, once convolutional and fully connected layers have learned hidden features. According to extensive tests, DCCR performs better than conventional collaborative filtering techniques, especially when dealing with sparse data, which highlights its potential for use in e-commerce and content streaming services.