Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Science

Program

Computer Science

Supervisor

Daley, Mark

Abstract

Temporal networks exhibit evolving patterns of connectivity, making the study of community organization a critical component of their analysis. Communities within a network are characterized by entities displaying higher interconnectivity within the community than outside it. In temporal networks, these communities can change over time, requiring a more nuanced analysis compared to static networks. However, there is a scarcity of statistical tools designed to quantify and compare dynamic community organization in temporal networks. In this study, we introduce a measure tailored for this purpose. We use this measure to discriminate between functional brain networks in which subjects were tasked with recollecting different emotional states under fMRI.

Summary for Lay Audience

Networks are a fundamental part of the world, describing connections in systems like social groups, power grids, and even the human brain. Unlike fixed networks, some networks change over time, with groups or "communities" within them forming and shifting dynamically. Studying these changes is important but challenging, as most existing tools are designed for networks that remain static.

To address this, a method was developed to analyze the dynamics of communities in evolving networks. This method uses mathematical techniques to capture how different parts of a network interact across both space and time, providing a detailed understanding of its behavior as it changes.

The method was demonstrated by applying it to brain networks using data from fMRI scans, examining how different regions of the brain connect and reorganize during the recollection of emotional experiences. This application highlights the versatility of the approach for studying dynamic networks in complex systems, showing its potential for broad use in fields where network evolution plays a key role.

Download

Share

COinS