In this chapter, we have learned different approaches for recommender systems, such as similarity-based, content-based, collaborative filtering, and hybrid. Additionally, we discussed the downsides of these approaches. Then we implemented an end-to-end book recommendation system, which is a model-based recommendation with Spark. We have also seen how to interoperate between ALS and matrix factorization to efficiently handle a utility matrix.
In the next chapter, we will explain some basic concepts of deep learning (DL), which is one of the emerging branches of ML. We will briefly discuss some of the most well known and widely used neural network architectures. Then, we will look at various features of DL frameworks and libraries.
Then we will see how to prepare a programming environment, before moving on to coding with some open source DL libraries, such as Deeplearning4j...
