An Enterprise Knowledge Graph (EKG) contains business objects and topics that are closely linked, classified, semantically enriched, and connected to existing data and documents.
An EKG typically consists of three pillars:
- A domain model—consisting of a conceptual and linguistic model— that is created and maintained by knowledge engineers and subject-matter experts with the help of machine learning algorithms. This serves as a common structural interface for all of your data and also forms the foundation necessary to create "data graphs" automatically.
- A data graph that represents intelligent multilateral relationships within your databases, content and document repositories. This acts as an additional virtual data layer connecting all of your data, no matter the scale and regardless of whether it is structured or unstructured.
- A user graph as part of the data graph containing users’ semantic profiles. This can be partly or automatically derived from user behavior. The user graph also links users with each other and with knowledge and data objects in a targeted manner.