Semi-metric networks for recommender systems

TL;DR

This paper explores the use of semi-metric properties of weighted co-occurrence graphs to improve recommender systems, demonstrating enhanced performance on benchmark datasets by leveraging semi-metric edges.

Contribution

It introduces a novel approach that utilizes semi-metric graph behavior for recommendations and analyzes its relation to fuzzy graph transitive closure.

Findings

Semi-metric edges improve recommendation accuracy.

Method outperforms traditional item- and user-based systems.

Enhanced recommendations on Movielens benchmark.

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.