Presentation Type

Presentation

Streaming Media

Location

Schimmel/Conrades Science Center 163

Start Date

18-4-2018 6:00 PM

End Date

18-4-2018 6:05 PM

Disciplines

Computer Sciences | Neuroscience and Neurobiology | Physics | Political Science

Keywords

Physics; Politics; Computer-Science; Congress; Network Theory; Directed Networks; Influence

Abstract

Research into the identification of influential nodes specifically with regards to weighted, directed networks has been lacking throughout the lifetime of Network Theory as a whole. This research project seeks to propel the field forward through by devising an algorithm aimed at identifying influential nodes through the use of probability propagation models of information transfer through various real-world networks. The networks discussed are a developed test-network using the Price Model of citation network growth, the neuronal connectivity network of the flatworm C. elegans, and Congressional co-sponsorship networks of the USA’s 110th House and Senate. Rankings of influence of each node, as determined by the developed algorithm, were then compared to ground truth rankings of influence determined by Susceptible-Infected-Recovered (SIR) model simulations via the Kendall’s Tau-b statistical measurement, so as to determine the reliability and accuracy of the developed algorithm’s probabilistic approach to influence. To this end, the research finished off by examining how the developed algorithm’s results compare to that of well-known influence calculation mechanisms, such as Google’s PageRank.

Project Origin

Summer Research Opporutnity

Faculty Mentor

Christian Fink

Share

COinS
 
Apr 18th, 6:00 PM Apr 18th, 6:05 PM

Identifying Influentials in Directed Networks

Schimmel/Conrades Science Center 163

Research into the identification of influential nodes specifically with regards to weighted, directed networks has been lacking throughout the lifetime of Network Theory as a whole. This research project seeks to propel the field forward through by devising an algorithm aimed at identifying influential nodes through the use of probability propagation models of information transfer through various real-world networks. The networks discussed are a developed test-network using the Price Model of citation network growth, the neuronal connectivity network of the flatworm C. elegans, and Congressional co-sponsorship networks of the USA’s 110th House and Senate. Rankings of influence of each node, as determined by the developed algorithm, were then compared to ground truth rankings of influence determined by Susceptible-Infected-Recovered (SIR) model simulations via the Kendall’s Tau-b statistical measurement, so as to determine the reliability and accuracy of the developed algorithm’s probabilistic approach to influence. To this end, the research finished off by examining how the developed algorithm’s results compare to that of well-known influence calculation mechanisms, such as Google’s PageRank.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.