With the huge amount of digital information available on the internet, it becomes a challenge for users to access items efficiently. Recommender systems are information filtering systems that deal with the problem of digital data overload to pull out items or information according to the user's preferences, interests, and behavior, as inferred from previous activities.
In this chapter, we will cover the following topics:
- Introducing recommender systems
- Latent factorization-based collaborative filtering
- Using deep learning for latent factor collaborative filtering
- Using the restricted Boltzmann machine (RBM) for building recommendation systems
- Contrastive divergence for training RBMs
- Collaborative filtering using RBMs
- Implementing a collaborative filtering application using RBMs
