English

Semi-metric networks for recommender systems

Information Retrieval 2012-09-11 v1 Statistical Mechanics Social and Information Networks

Abstract

Weighted graphs obtained from co-occurrence in user-item relations lead to non-metric topologies. We use this semi-metric behavior to issue recommendations, and discuss its relationship to transitive closure on fuzzy graphs. Finally, we test the performance of this method against other item- and user-based recommender systems on the Movielens benchmark. We show that including highly semi-metric edges in our recommendation algorithms leads to better recommendations.

Keywords

Cite

@article{arxiv.1209.1719,
  title  = {Semi-metric networks for recommender systems},
  author = {Tiago Simas and Luis M. Rocha},
  journal= {arXiv preprint arXiv:1209.1719},
  year   = {2012}
}
R2 v1 2026-06-21T22:01:55.261Z