Ranking RDF Instances in Degree-decoupled RDF Graphs

TL;DR

This paper introduces three new importance measures for RDF graph nodes that are decoupled from node degree, and compares them with traditional centrality measures using IMDb data.

Contribution

It proposes three novel importance measures tailored for degree-decoupled RDF graphs and evaluates their effectiveness against existing measures.

Findings

InfoRank measures outperform traditional PageRank in degree-decoupled RDF graphs.

The proposed measures provide more accurate importance scores in specific RDF graph structures.

Experimental results demonstrate the relevance of degree-independent importance metrics.

Abstract

In the last decade, RDF emerged as a new kind of standardized data model, and a sizable body of knowledge from fields such as Information Retrieval was adapted to RDF graphs. One common task in graph databases is to define an importance score for nodes based on centrality measures, such as PageRank and HITS. The majority of the strategies highly depend on the degree of the node. However, in some RDF graphs, called degree-decoupled RDF graphs, the notion of importance is not directly related to the node degree. Therefore, this work first proposes three novel node importance measures, named InfoRank I, II and III, for degree-decoupled RDF graphs. It then compares the proposed measures with traditional PageRank and other familiar centrality measures, using with an IMDb dataset.