PROVDM

The PROV Data Model, commonly abbreviated PROV-DM or PROVDM, is a World Wide Web Consortium (W3C) standard for representing provenance information. It provides a formal data model for describing the origins, history, and lineage of data and artifacts.

The model centers on three core node types: Entity, Activity, and Agent. Entities are data items or artifacts; Activities are processes or actions that use and generate entities; Agents are responsible actors. Relationships connect these nodes: wasGeneratedBy (an entity generated by an activity), used (an activity used an entity), wasAssociatedWith (an activity was associated with an agent), wasAttributedTo (an entity was attributed to an agent), and wasDerivedFrom (an entity derived from another). The model supports time attributes and optional metadata.

PROV-DM is designed to be domain-agnostic and to interoperate with multiple serializations, including PROV-N (a human-readable notation), PROV-JSON, and PROV-XML. It also aligns with the PROV-O ontology for semantic web use and other related specifications. It is widely used in scientific workflows, data management, and auditing to enable reproducibility and accountability.

The standard emphasizes extensibility and compatibility with existing data ecosystems, allowing organizations to record provenance alongside datasets, workflows, and publications. It serves as a foundational layer for provenance representation and interoperability among systems that exchange metadata.