Semantic Relatedness in DBpedia: A Comparative and Experimental Assessment

The evaluation of semantic relatedness between web resources is still an open chal-lenge. This paper focuses on knowledge-based methods that provide an alternative tocorpus-based approaches and generally rely on the availability of knowledge graphs.In particular, we selected 10 methods from the existing literature, organized byadjacent resources, triple patterns, and triple weights-based methods. They were im-plemented and evaluated using DBpedia as a reference RDF knowledge graph. SinceDBpedia is under constant development, the experimental results of these proposalsare not comparable in the literature. For this reason, in this work, these methodswere experimented with the same DBpedia release and 14 well-known golden datasetsat once. Based on this experimentation, the weighting of RDF triples combined withthe evaluation of all the directed paths linking the compared resources is the beststrategy for computing semantic relatedness in DBpedia