A property graph is a type of graph model where relationships not only are connections but also carry a name (type) and some properties. A property graph excels at showing connections among data scattered across diverse Data Architectures and data schemas. They provide a richer view on how data can be modeled over many different databases as well as how different kinds of metadata relate. Property graphs also show data dependencies not seen in a relational database schema or through other tools.
The International Standards Organization (ISO) is developing Property Graph Language (GQL) for querying property graphs created by different organizations.
“An extension of the directed graph paradigm with the concept of properties located on nodes, and also potentially on relationships.”(Thomas Frisendal)
“A logical model that is not a Resource Description Framework (RDF) but has nodes and properties that connect nodes and also store attributes (scalar values).” (Forbes)
A graph model “consisting of entities, often called nodes, and links between them, often called relationships. Nodes and relationships can also contain properties and attributes.” (Open Data Science (ODSC) Team)
A “form of a graph that some popular tools use to manage graph data.” (Big Data Quarterly (BDQ)
A graph like a semantic graph but it differs in that it “treats the edge as something unique based upon the surrounding nodes.” (Forbes)
A graph that allows “properties (key/value pairs) to be associated with both nodes and links in directed graphs.” (W3C)
