Loading...
Please wait, while we are loading the content...
Similar Documents
Analysis of a Voting Method for Ranking Network Centrality Measures on a Node-aligned Multiplex Network
| Content Provider | Semantic Scholar |
|---|---|
| Author | Wilkinson, Kyle S. |
| Copyright Year | 2018 |
| Abstract | In Joint Concept: Human Aspects of Military Operations (JC-HAMO), the need for identifying critical actors within a target network is clearly identified as a precursor to successfully influencing decision making and operational outcomes. JC-HAMO seeks methods to identify critical actors within the context of multiple types of networks and over a period of time. This problem can be approached structurally using a time-stamped multilayer network. One method of identifying critical actors in a single layer—fullyaggregated—network involves ranking actors in order of importance by some set of network measures. This thesis explores a method for extending such a ranking of critical actors into a multilayer network context. Specifically, it borrows and applies a methodology from the field of electoral systems to the problem of ranking actors based on a set of rankings for each layer. The Schulze method—a deterministic voting methodology based on a modified shortest path algorithm—is examined and its performance is assessed through statistical comparison with identified alternative approaches and baseline rankings. Potential advantages and limitations are identified as well as a method for increasing its robustness when the networks of interest contain many isolated components. This is done by adopting a secondary weighting scheme. As a corollary study, an information-theoretic multilayer network layer-reduction heuristic is explored and the resulting rankings on the reduced multilayer network are compared with those of the full multilayer network and those of the corresponding fully-aggregated single layer network. A tertiary effort compares two |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://scholar.afit.edu/cgi/viewcontent.cgi?article=2867&context=etd |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
