Research Colloquium Presentation by Nikolaos Karalis

3 years ago

Knowledge Graphs

Authors: Aidan Hogan, Eva Blomqvist, Michael Cochez, Claudia d'Amato, Gerard de Melo, Claudio Gutierrez, José Emilio Labra Gayo, Sabrina Kirrane, Sebastian Neumaier, Axel Polleres, Axel-Cyrille Ngonga Ngomo, Sabbir M. Rashid, Anisa Rula, Lukas Schmelzeisen, Juan Sequeda, Steffen Staab, Antoine Zimmermann

This paper covers in detail many research topics and aspects of knowledge graphs (KGs). In this presentation we focus on the topic Data Graphs and briefly cover the topics Schema and Identity.

In the section dedicated to the topic of Data Graphs, the authors formally define the different graph data models that are used for the representation of KGs. The most prominent models are the directed edge-labelled graph model and the property graph model. The former has been adopted by the Resource Description Framework (RDF) and the semantic web community, whereas the latter is used by graph databases providers. Furthermore, the authors present the fundamentals of graph query languages (i.e., graph patterns and navigational graph patterns), which are shared across different graph data models.

There are different types of schemata that can be used in the context of KGs. A semantic schema can be used in order to define the meaning of high-level terms, such as classes (e.g., classes), whereas a validating schema can be used to define constraints on a data graph. Last but not least, the authors stress the importance of associating entities of a KG with an identifiers and present identification examples in the context of RDF.

The slides of the presentation can be found here