|Date:||2022, April 19|
|Time:||10:30 a. m.|
|Title:||Knowledge Graphs and their Embeddings|
Knowledge engineering and reasoning are well-established topics of research in the domains of artificial intelligence and logic. Knowledge graphs (KG) are effective tools for representing structural relationships between concepts or real-world entities. One significant difficulty with knowledge graphs is that they are often incomplete. While deductive reasoning methods based on symbolic logic are interpretable, they cannot cope with uncertainties and missing information within the knowledge graphs. To address this problem, more recent approaches have focused on representing entities and relations in continuous vector spaces while keeping their semantics. These vector representations are easy to manipulate and allow the inference of additional facts. In this presentation, we will discuss existing knowledge graph embedding approaches and the types of relations that they can model.