Network Elucidation Template: A framework for human-guided network inference

Leo Lopes, Jay Konieczka, Victor Foulk, Parker Antin

Research output: Contribution to journalArticle

Abstract

Network elucidation is the problem of inferring all parameters of a network from a subset of those parameters. We introduce the Network Elucidation Template (NET), which provides a framework upon which algorithms for such problems can be built. NET algorithms take advantage of novel methods for collaboration between human operators and computers. They use visualizations of the peculiar structures that appear in optimal solutions to aid the parameter search. By design, NET is at a high enough level of abstraction to describe a class of algorithms, as opposed to a single algorithm. Given a problem, and the structure of that problem, an effective instantiation of the template into an algorithm can be created. We describe one such instantiation: using a network flow framework to implement a NET algorithm for uncovering smuggling networks; as well as the general template.

Original languageEnglish (US)
Pages (from-to)680-690
Number of pages11
JournalComputers and Industrial Engineering
Volume58
Issue number4
DOIs
StatePublished - May 1 2010

Keywords

  • Human-guided search
  • Optimization
  • Visualization

ASJC Scopus subject areas

  • Computer Science(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'Network Elucidation Template: A framework for human-guided network inference'. Together they form a unique fingerprint.

  • Cite this