Mapping Management and Expressive Ontologies in Ontology-Based Data Access
Diego Calvanese, Benjamin Cogrel and Guohui Xiao
Ontology-Based Data Access (OBDA) has become a popular paradigm for overcoming the typical difficulties in accessing data stored in different kinds of legacy sources, by leveraging a conceptual representation of such data provided in terms of an ontology. OBDA relies on bridging database systems and technologies developed in the areas of Knowledge Representation and in the Semantic Web. The main mechanism for establishing such a bridge are RDF-to-database mappings, which allow one to create virtual views characterizing the content of the data sources and directly relate them to the terms in the ontology. In this tutorial: (i) We provide a general introduction to the principles and basic technologies for OBDA, relying on RDFS, the OWL~2~QL profile, and on the transformation of SPARQL into native database queries. (ii) We let participants practice such technologies in a hands-on session using Protege and the OBDA system Ontop. (iii) We provide more insights into the theoretical foundations of OBDA, by analyzing query rewriting and the computational impact of ontologies on query answering. (iv) We continue with a further hands-on session on mapping engineering and deployment of Ontop as a SPARQL endpoint. (v) We provide an overview on the latest advancements in OBDA concerning non-relational (NoSQL) datasources, and approximation techniques.
Modeling, Generating and Publishing knowledge as Linked Data
Anastasia Dimou, Pieter Heyvaert and Ruben Verborgh
Knowledge Acquisition and Modeling are important in a world with large heterogenous data sources. This process of extracting, structuring, and organizing knowledge from one or multiple data sources is required to construct knowledge-intensive systems and services for the Semantic Web. This way, the processing of large and originally semantically heterogeneous data sources is enabled and new knowledge is captured. Thus, offering existing data as Linked Data increases its shareability, extensibility and reusability. However, using Linking Data, as a means to represent knowledge, has proven to be easier said than done. During this tutorial, we will elaborate the importance of semantically annotating data and how existing technologies facilitate their mapping to Linked Data. We will introduce the [R2]RML, language(s) to generate Linked Data derived from different heterogeneous data sources, e.g., tabular data in databases, data in XML published as Open Data or data in JSON derived from a Web API. More, we will support non-Semantic Web experts to annotate their data with the RMLEditor. Through the tool’s innovative user interface all underlying Semantic Web technologies are invisible to the end users. Last, we will show how to easily publish Linked Data with LDF. In the end, participants, independently of their knowledge background, will have model, annotate and publish some data on their own!