Visual Analysis and Querying of Large Knowledge Graphs
Session Overview
Grasping and querying large Knowledge Graphs (KGs) is challenging. I will demonstrate tooling for RDF as well as LPG graph data that uses visual aggregation to enable exploration and querying of large KGs, from the meta level down to node and relationship details without the hairball problem of graph visualization.
Using production data from an industry project, I will share experiences with our codeless query approach that can guide users in developing meaningful queries without any prior idea of the data or any SPARQL or Cypher language skills to:
investigate not only the structure but also the proportions of millions of nodes in the KG,
find data flaws, identify data outliers or missing connections,
build complex queries including sophisticated filtering and path expressions just by visual exploration,
refine the graph schema with new classes/labels on the fly while browsing the graph.
Target Audience
Data stewards, data analysts, data owners, data domain experts or any other user with a need for
graph understanding
graph data quality analysis
query building
Desired Feedback
User feedback about the tool set or exchange about alternative approaches / workflows etc.
Speaker Details
Thorsten Liebig is co-founder of derivo, who is supporting companies and organizations to reveal and exploit the knowledge in their data sources by means of semantic technologies. As a consultant he has contributed to KG projects at Festo, Schaeffler, Springer Nature or Bertelsmann. Thorsten was responsible co-author of OWLlink, an official W3C member submission.
Talk Title
Visual Analysis and Querying of Large Knowledge Graphs
Session Overview
Grasping and querying large Knowledge Graphs (KGs) is challenging. I will demonstrate tooling for RDF as well as LPG graph data that uses visual aggregation to enable exploration and querying of large KGs, from the meta level down to node and relationship details without the hairball problem of graph visualization.
Using production data from an industry project, I will share experiences with our codeless query approach that can guide users in developing meaningful queries without any prior idea of the data or any SPARQL or Cypher language skills to:
Target Audience
Data stewards, data analysts, data owners, data domain experts or any other user with a need for
Desired Feedback
User feedback about the tool set or exchange about alternative approaches / workflows etc.
Speaker Details
Thorsten Liebig is co-founder of derivo, who is supporting companies and organizations to reveal and exploit the knowledge in their data sources by means of semantic technologies. As a consultant he has contributed to KG projects at Festo, Schaeffler, Springer Nature or Bertelsmann. Thorsten was responsible co-author of OWLlink, an official W3C member submission.
Recording Consent
YES
Workshop Potential
NO
Visual Aid for Social Media
Further information and links: