Definition
A graph data model structures data into interconnected objects that form a graph. It is useful:
- for highly interconnected data
- particularly popular in online social networks
Two entities of a graph data model:
- objects: nodes that have a unique identifier (ID) and a type (OTYPE)
- associations: edges, have a source object (ID1), a destination object (ID2) and a type (ATYPE)
Attributed graph In an attributed graph, both objects and associations may have properties, expresses as key-value pairs
Operations over graph database
- retrieval of objects and associations
- creation of objects and associations
- deletion of objects and associations
- mutation of objects and associations
Example
In a social media, objects or nodes can be represented by users, business, groups or photos, while associations may represents friendships, follower-follower relationships or “like” relationships
References
- Invariant detector finder over graph database, by (Marinescu, Parry, et al., 2017), Facebook