Table 1.

Comparison of the different data models

Property-based modelTimeline modelHierarchy model
ExpressivenessLow, simple property with dedicated datatype (only storage, no semantic information), supports time zones, the property is not unique, needs an additional property to resolve timestamp equality, the user must know how to interpret stored dataMedium, simple property with dedicated datatype in dedicated node, supports time zones, the node can be unique, semantic information: absolute temporal order given by directed edgesHigh, tree like structure with shared nodes, each node is unique, allows indication of timestamp precision, fine grained node fetching possible, no time zone support, the relative temporal order of nodes of the same level and sub tree by directed edges, requires directed edge to resolve timestamp equality
ComplexityLow, comparable complex as the equivalent query in a relational databaseMedium, the problem described in Section 7.1 forces usage of two different approaches for node access and matchingHigh requires complex queries due to different levels when searching specific nodes
SuitabilityDoes not exploit graph database specific functionalityBasic usage of nodes and directed edges by constructing a global timelineExtensive usage of nodes and edges, most information expressed by these elements

or Create an Account

Close Modal
Close Modal