Navigating Contextual Complexity with Graph Visualization
Within any of these navigation strategies , it is also possible to provide selective navigation based on the type of edge ( s ) to be navigated , or to specify the distance to navigate , in terms of either the number of connections ( hops ), or by somehow specifying the target objective of the navigation ( the node ( s ) to be reached ).
2.1 TOOLS FOR NAVIGATION AND INTERACTION
It is possible to categorize methods of user interaction as either acting on the data to be displayed ( selection , filtering , calculation ), or on the way that data is presented ( layout style , or other visual parameters ). [ 7 ] These tools work together , and how they work forms the backbone for how navigation is afforded within the user experience .
Control Method
Aggregation Abstraction Context Spatial Navigation
Data-level examples |
Statistical clustering , or domain-based aggregation .
Abstraction by query or through Graph AI summarization techniques .
|
Query-based context selection |
Location-based query , either geographic , or topological . |
Path query or path traversal algorithm . |
Display-level examples |
Grouping , nesting , combining . |
Applying attributebased , time-based , or statistical abstraction . |
Filter-based context visibility |
Projection to a map , or layout constraints . |
Follow links , find by search or selection . |
Table 2-1 : Sample of tools and approaches for context and complexity management .
3 SPECIAL CONSIDERATIONS FOR SYSTEM ENGINEERING
Model-based System Engineering ( MBSE ) is discipline that relies heavily on diagrams as both a canonical input mechanism and communication tool for collaboration . SysML v1 is a modeling system that utilizes diagram syntax as the declarative form of the model [ 8 ]. This mode of visual graphical editing can be natural , but it is also prohibitive in terms of the number of different diagrams that are required to communicate all aspects of the design or behavior of the system for all the stakeholders that demand customized diagrams to illustrate specific concerns .
Furthermore , it is up to the engineers modeling these systems to ensure that changes in the model are manually propagated to all the relevant diagrams . This can lead to diagrams that are missing elements or connections or contain other inconsistencies that are difficult to identify through manual review and may introduce errors . Undetected , errors like these can be propagated across multiple teams and product components . [ 9 ] The longer such errors remain undetected , the more likely they are to create defects in implementation that can become costly .
76 February 2025