Amazon.com Recommendations. Item-to-item Collaborative Filtering.

Amazon.com Recommendations. Item-to-item Collaborative Filtering.

Greg Linden, Brent Smith, and Jeremy York. 2003. (View Paper → )

At Amazon.com, we use recommendation algorithms to personalise the online store for each customer. The store radically changes based on customer interests, showing programming titles to a software engineer and baby toys to a new mother. The click-through and conversion rates — two important measures of Web-based and email advertising effectiveness — vastly exceed those of un-targeted content such as banner advertisements and top-seller lists

Amazon were pioneers of early recommender systems. The authors technique cleverly reduces compute required vs greedier methods. They navigate data sparsity issues. Much of the calculation to happens offline but recommendations can remain responsive in realtime.