Summary

In this chapter, we discussed building personalized product recommendation systems. First, we discussed how to identify interrelationships between products by analyzing which itemsets are frequently bought together. We covered how to conduct market basket analysis in Python using the Apriori algorithm and association rules. Then, we dove deep into one of the AI/ML-driven approaches to building recommendation systems. We saw how user-based and item-based collaborative filtering algorithms can be used to identify similar users or items and products that are frequently bought together. In turn, these findings can then be used to recommend certain products that other customers are highly likely to be interested in and purchase. Lastly, we discussed various other approaches that can be used for recommending products and content. We showed how combining non-AI/ML approaches can result in a more comprehensive and diverse recommendation system.

In the following chapter, we will...