An Interactive Perception Method Based Collaborative Rating Prediction Algorithm

To solve the rating prediction problems of low accuracy and data sparsity on different datasets, we propose an interactive perception method based collaborative algorithm named DCAE-MF, by fusing dual convolutional autoencoder (DCAE) probability matrix factorization (PMF). Deep latent representations users items are captured simultaneously DCAE deeply integrated with PMF to collaboratively make predictions known history users. A global multi-angle optimization learning is developed effectively optimize all parameters DCAE-MF. Extensive experiments performed seven real-world datasets demonstrate superiority DCAE-MF key metrics root mean squared error (RMSE) absolute (MAE).