Summary

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.

This is more or less the end of our journey toward developing an end-to-end project with Java. However, we are not done yet! In...