KG Relationships
Typed, directed edges connecting entities/concepts in knowledge graphs, enabling semantic reasoning (e.g., inference), provenance tracking, hierarchies, temporality, and attribution. These are core building blocks for KG design, often weighted (e.g., confidence 0.8) and extracted via AI techniques like relation extraction.
Examples from Diagram
The 10 relationships below exemplify a robust KG schema:
| Type | Semantics | Use Case Example |
|---|---|---|
| supports | Evidence agreement | Claim validation |
| contradicts | Evidence conflict | Contradiction flagging |
| targets | Influence or aims at | Goal/action planning |
| derived_from | Origin/derivation | Data provenance |
| lineage | Ancestry/heritage chain | Historical tracking |
| part_of | Hierarchical inclusion | Part-whole decomposition |
| preceded_by | Temporal predecessor | Event sequencing |
| followed_by | Temporal successor | Event sequencing |
| authored_by | Creator attribution | Credit/intellectual property |
| tagged_with | Categorization/labeling | Indexing/discovery |
– Source: https://youtu.be/z02Y-1OvWSM?si=Xi9Q_R7RwpElYCHIKG Relationships Diagram
Source: https://youtu.be/z02Y-1OvWSM?si=Xi9Q_R7RwpElYCHI
Link to original
Role in KGs
- AI Symbiosis: Fuel Graph Data Science predictions, explainable reasoning.
- Design Principles: Extend with domain ontologies (e.g., RDF, schema.org).