Navigating Contextual Complexity with Graph Visualization
prescribes all the things , and patterns of connected things , that it is possible to represent with that meta-model . The simple model depicted in Figure 4-1 includes one type of node or vertex in the graph , called a Device , which has several attributes . It includes one edge type , Connection , which also has attributes . Each attribute has a defined data type . Rules or constraints could be added to restrict the string values allowed for specific attributes , or prevent instances described by the model from allowing a Device to have a connection to itself , for example .
Figure
4-1 : Simple meta model with one node type and one edge type .
Tools for interacting with a graph schema , whether an ontology or property graph , should also afford users with the utility to analyze the schema at various levels of detail , to support discovery of unique properties or patterns within the schema , and to identify inconsistencies or gaps .
Industry ontologies are an example of large and complex graphs that can benefit from multiple modes of visualization that are not always readily offered in schema design interfaces . Graphbased systems can help to both illustrate and to enforce common patterns in collaborative or crowd-sourced information models , with less overhead in review and analysis of new submissions to the model .
The following examples are derived from the last freely available draft of the Industry Foundation Classes Model ( IFC4x3 ), the foundation of the Building Information Modeling standard ISO 19650- 1:2018 , maintained by buildingSMART International , and accessible from buildingSMART . org . The RDF ontology is generated from the native form of the IFC model , which is STEP ( ISO 10303- 21 ). All the following visualizations were produced with Tom Sawyer Perspectives .
Journal of Innovation 79