Public defence in the field of Mathematics and Statistics, M.Sc. Ryan Wood

Title of the thesis: Non-backtracking Centrality Measures and Beyond

Thesis defender: Ryan Wood
Opponent: Senior Lecturer Philip Knight, University of Strathclyde, United Kingdom
Custos: Associate Professor Vanni Noferini, Aalto University School of Science

Networks are fundamental mathematical structures that appear in a range of fields and applications. One of the most fundamental questions one can ask about a network is which nodes are most influential. Centrality measures are functions which assign to each node a non-negative value indicative of their importance within the network and are ubiquitous in many areas of study

Centrality measures based on non-backtracking walks have been shown to yield concrete benefits over popular walk-based centralities, such as Katz centrality. However, the computational cost of such non-backtracking centralities can be prohibitively high and the classes of graphs to which they can be applied is limited.

The research presented in this doctoral thesis seeks to overcome these challenges and facilitate the use of non-backtracking centrality measures as a tool for analyzing time-evolving and/or weighted networks.    

Thesis available for public display 10 days prior to the defence at Aaltodoc.