In this chapter, we saw how to develop a movie recommendation system using FMs, which are a set of algorithms that enhance the performance of linear models by incorporating second-order feature interactions that are absent in matrix factorization algorithms in a supervised way.
Nevertheless, we have seen some theoretical background of recommendation systems using matrix factorization and collaborative filtering before diving into the project's implementation using RankSys library-based FMs. Due to page limitation, I didn't discuss the library more extensively. However, readers are suggested to take a look athe API documentation on GitHub at https://github.com/RankSys/RankSys.
This project not only covers movie rating prediction by individual users but also discusses ranking prediction, too. Consequently, we also used FMs for predicting the ranking of movies.
...
