University/Organization
University of Detroit Mercy
Title
The Evaluation of Similarity Metrics in Collaborative Filtering Recommenders
Synopsis
We evaluate multiple similarity measures in a traditional collaborative filtering process. We also consider combinations of complementary measures, especially in edge cases when one of them falls short, e.g., a user with uniform ratings. We examine prediction accuracy, classification accuracy, confusion statistics, and actual/predicted distribution compatibility to find the best way to quantify vector similarity.