Embedding Models for Knowledge Graphs Induced by Clusters of Relations and Background Knowledge

Embedding models have been successfully exploited for predictive tasks on Knowledge Graphs (KGs). We propose TRANSROWL-HRS, which aims at making KG embeddings more semantically aware by exploiting the intended semantics in the KG. The method exploits schema axioms to encode knowledge that is observed as well as derived by reasoning. More knowledge is further exploited by relying on a successive hierarchical clustering process applied to relations, to make use of the several semantic meanings that the very same relation may have. An experimental evaluation on link prediction and triple classification tasks proves the improvement yielded by the proposed approach (coupled with different optimizers) compared to some baseline models.