In this chapter, you learned about the basic concept of recommendation engines, the differences between collaborative and content-based filtering, and how to use Apache Mahout, which is a great basis for creating recommenders, as it is very configurable and provides many extension points. We looked at how to pick the right configuration parameter values, set up rescoring, and evaluate the recommendation results.

With this chapter, we have completed our overview of the data science techniques that are used to analyze customer behavior, which started with customer relationship prediction in Chapter 4, Customer Relationship Prediction with Ensembles, and continued with affinity analytics in Chapter 5, Affinity Analysis. In the next chapter, we will move on to other topics, such as fraud and anomaly detection.