Rethinking Recommender Systems: Cluster-based Algorithm Selection

TL;DR

This paper demonstrates that clustering users before selecting recommendation algorithms significantly improves performance across multiple datasets, with substantial gains in nDCG@10 scores.

Contribution

It introduces a comprehensive study on cluster-based algorithm selection, exploring various clustering methods and recommendation algorithms to optimize performance.

Findings

Cluster-based selection improves nDCG@10 by up to 360%.

Effective across eight diverse datasets.

Shows the importance of user clustering in recommendation systems.

Abstract

Cluster-based algorithm selection deals with selecting recommendation algorithms on clusters of users to obtain performance gains. No studies have been attempted for many combinations of clustering approaches and recommendation algorithms. We want to show that clustering users prior to algorithm selection increases the performance of recommendation algorithms. Our study covers eight datasets, four clustering approaches, and eight recommendation algorithms. We select the best performing recommendation algorithm for each cluster. Our work shows that cluster-based algorithm selection is an effective technique for optimizing recommendation algorithm performance. For five out of eight datasets, we report an increase in nDCG@10 between 19.28% (0.032) and 360.38% (0.191) compared to algorithm selection without prior clustering.