Deep Learning for Recommender Systems

Addressing Item Cold Start Problem in Collaborative Filtering-Based Recommender Systems Using Auxiliary Information

“…In recent years, modeling user behavior using classical machine learning techniques has become a challenge with the large amount of information available for recommender systems and the complexity of user interactions with the system [49]. Thus, deep learning techniques have been increasingly used in recommender systems as they enable processing unstructured data and inferring hidden patterns of user preferences and item representation [30,49].…”

“…In recent years, modeling user behavior using classical machine learning techniques has become a challenge with the large amount of information available for recommender systems and the complexity of user interactions with the system [49]. Thus, deep learning techniques have been increasingly used in recommender systems as they enable processing unstructured data and inferring hidden patterns of user preferences and item representation [30,49]. Also, deep learning techniques are able to consider descriptive information (such as text, images, audio, and video) about users and items available from various sources, to create a more reliable and accurate user model [49].…”

“…Thus, deep learning techniques have been increasingly used in recommender systems as they enable processing unstructured data and inferring hidden patterns of user preferences and item representation [30,49]. Also, deep learning techniques are able to consider descriptive information (such as text, images, audio, and video) about users and items available from various sources, to create a more reliable and accurate user model [49]. Deep learning has been successfully applied in fields as diverse as movie, music, image or quote recommendation.…”

“…As the architecture of a neural network has a decisive influence on the learning model performance, improving the construction of the neural system, for example by adding hidden layers to the neural network, may help the multitask learning. Also, implementing recommendation algorithms through deep neural networks involves high computational cost that should be reduced using more efficient algorithms [12,49]. The big amount of data required for the learning process is another disadvantage of deep learning techniques, since it is not possible to produce accurate recommendations when there is not enough data available [49].…”

“…As stated in [39,49], the cold start problem is one of the research challenges of recommender systems. The problem has caught the attention of many researchers [6,17] because the ability of a recommender system to produce relevant recommendations often comes up against the long-tail distribution of the number of ratings per user.…”