Heiko Paulheim.
Abstract: In the recent years, vector space embeddings of semantic web knowledge graphs — i.e., projections of a knowledge graph into a lower-dimensional, numerical feature space (a.k.a. latent feature space) — have been shown to yield superior performance in many fields, including relation prediction, recommender systems, or the enrichment of predictive data mining tasks. At the same time, such a representation is as non-semantic as possible, hence, the results are most often not interpretable, and it is hard to obtain a justification for a prediction, e.g., an explanation why an item has been suggested by a recommender system. In this paper, we make a claim for semantic embeddings and discuss a possible (yet unvalidated) approach for their construction.
Keywords: Knowledge Graphs; Embeddings; Representation Learning