OntoRipple: Making waves in the knowledge graph lifecycle

Knowledge Graphs are commonly organised according to the structure of existing ontologies, which define the concepts, relations, and restrictions of the domain of the KG. There are ontology-dependent assets that guide how data from heterogeneous sources is integrated, transformed, validated, and exploited in the KG, such as mapping rules and validation constraints. As ontologies evolve over time, these changes must be consistently reflected in the dependent assets, ensuring that the resulting KG remains aligned with the updated ontology. While ontology evolution has been widely studied, the propagation of changes to dependent artifacts remains an open challenge, requiring manual effort that makes the process slow, error-prone, and costly. In this paper, we present OntoRipple, a set of algorithms integrated into a tool that automates the propagation of ontology changes into RML mappings and SHACL shapes to construct and validate Knowledge Graphs, ensuring consistency with the evolving ontology in a fully declarative workflow.

keywords: Knowledge Graphs, Ontology Evolution, Mapping Rules, Impact Assessment, Validation rules