A New Hybrid Deep Learning Model based-Recommender System using Artificial Neural Network and Hidden Markov Model
Choukri Djellali, Mehdi Adda · Procedia Computer Science · 2020
Recommendation systems based on Deep Learning have recently led to significant progress in different application domains. Most models are influenced by hyperparameter optimization or tuning, the stability of training, and architecture configuration. Hence, in the present study, we introduced a Deep Learning model, which is named RHMM, for recommender systems, by using the Hidden Markov Model and Artificial Neural Networks. The model selection technique is applied to optimize the bias-variance tradeoff of the expected prediction. The model aggregation technique is used to improve the robustness and accuracy of training. Experiment results showed that our Deep Leaning model led to significant improvement over benchmarking models.