Summary

This chapter dealt with the topic of the recommender systems that are ubiquitous on our daily journeys online. Compared to the previous chapters, we didn’t perform any classification tasks; instead, we focused on the most noteworthy techniques in ML for implementing recommender systems. Utilizing a corpus of Amazon reviews, we tried to elicit customized suggestions for music titles.

To wrap up, in the first part of the chapter, we performed the necessary data cleaning to eliminate corrupted data that would affect the quality of the developed systems. Then, we manipulated the dataset to make it suitable for the analysis that followed. We also enhanced our arsenal of data visualization methods with new types of plots.

In the second part, we attacked the problem by focusing on the properties of products or customer ratings. We then detailed the suitable methods for both cases and implemented various recommenders. Simultaneously, we broadened our coverage of dimensionality...