Provenance for knowledge graphs: data, queries, and beyond

Graphs are a popular way of representing knowledge in a machine-interpretable format. In particular knowledge graphs are applied in diverse use cases that span over natural language processing, fact checking, recommender systems, semantic data lakes, and semantic data warehousing. Whereas the versatility of graph structures represents advantages when creating and publishing data, it results in challenges for scalable graph data management and querying along the value chain. One of these challenges is the support of provenance, which has so far been mostly neglected despite being essential to building trustworthy and explainable systems. After giving a short introduction, I will present two of our works on provenance support for RDF knowledge graphs: one uses query rewriting to compute how-provenance for SPARQL queries without requiring customized extensions to the underlying triple store. The other one shows how provenance information on data level can be exploited for OLAP-style SPARQL queries.

Short bio Katja is a professor in the Department of Computer Science at Aalborg University, where she is leading the Data, Knowledge, and Web Engineering group. Prior to joining Aalborg University, she was a postdoc at the Max Planck Institute for Informatics in Saarbrücken, Germany, and earned her PhD in Computer Science from Ilmenau University of Technology, Germany. Her work is rooted in databases and graph technologies and spans theory, algorithms, and applications of data science and knowledge engineering including knowledge graph management, querying, and analytics. She has co-authored more than 100 peer-reviewed scientific publications and regularly serves as a reviewer for a broad range of conferences and journals. She has served in many different organizational roles for international conferences, such as VLDB, SIGMOD, EDBT, TheWebConf/WWW, and ISWC, incl. program co-chair roles for EDBT 2023 and ESWC 2021.