Bias in Knowledge Graphs -- an Empirical Study with Movie Recommendation and Different Language Editions of DBpedia

TL;DR

This paper empirically investigates how using different language editions of DBpedia affects bias and performance in content-based movie recommendation systems, revealing significant variations.

Contribution

It is the first to analyze the impact of different language editions of DBpedia on bias and recommendation quality in a fixed recommendation strategy.

Findings

Different language editions lead to biased recommender systems.

Performance varies across recommendation fields depending on the knowledge graph used.

Bias and performance are significantly affected by the choice of knowledge graph.

Abstract

Public knowledge graphs such as DBpedia and Wikidata have been recognized as interesting sources of background knowledge to build content-based recommender systems. They can be used to add information about the items to be recommended and links between those. While quite a few approaches for exploiting knowledge graphs have been proposed, most of them aim at optimizing the recommendation strategy while using a fixed knowledge graph. In this paper, we take a different approach, i.e., we fix the recommendation strategy and observe changes when using different underlying knowledge graphs. Particularly, we use different language editions of DBpedia. We show that the usage of different knowledge graphs does not only lead to differently biased recommender systems, but also to recommender systems that differ in performance for particular fields of recommendations.