Please use this identifier to cite or link to this item:
Title: Complex network techniques for discovering structures in functional data
Authors: Goh, Woon Peng
Keywords: DRNTU::Science::Physics
DRNTU::Engineering::Computer science and engineering::Data
Issue Date: 2017
Source: Goh, W. P. (2017). Complex network techniques for discovering structures in functional data. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: Complex systems are systems with many interacting components that give rise to their rich dynamical behaviours. Complex networks provide a powerful and versatile perspective on complex systems, allowing us to define aspects such as hierarchy, flows, communities, and causality in terms of network topologies. In this thesis, we adopt such techniques of network analysis to interpret, represent, and reveal the internal organizations of three real-world complex data sets. Specifically, two examples used networks to represent and understand the dynamical processes of, respectively, teaching and problem-solving, and the third took a network approach to elucidate the syntactic organization of human language. In these examples, we use both existing and original techniques to infer networks from data, to filter dense networks into their most essential components, and to describe topological features of the network that correspond to the dynamics of their underlying systems. We show how network visualizations allow for an intuitive interpretation of complex dynamical data and the straightforward identification of generative drivers of dynamical sequences. Beyond visual interpretations, quantitative descriptions of the networks such as node betweenness and motif densities also reveal the internal logic of the systems.
DOI: 10.32657/10356/71754
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:IGS Theses

Files in This Item:
File Description SizeFormat 
Thesis.pdf6.76 MBAdobe PDFThumbnail

Page view(s)

Updated on Nov 30, 2020

Download(s) 50

Updated on Nov 30, 2020

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.