Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.756068
Title: Influence in economic and political systems : a network scientific approach
Author: Gurciullo, S. V.
ISNI:       0000 0004 7429 024X
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Complex social systems strive by exchanging information and resources. By means of the exchange, some actors in the system are able to at least partially determine the behaviour of another actor, thereby influencing it. Both the information exchange process and the degree of actors’ influence are latent, unobserved phenomena in many instances of real-world systems. This thesis presents a framework that intends to unearth the two hidden properties. It does so by introducing a Network Inference and Influence Framework (NIIF), which makes use of graph-based methods to derive a latent network in a social system, and measure the influence of its elements. The framework is applied on three case studies where the latency problem translates into research questions with importance for public policy making. The first case study uses NIIF to estimate the latent network of interdependency across financial institutions, and measures the extent to which a bank may negatively influence the system after an economic distress. In the second case study, a network of information diffusion is extracted from House of Commons parliamentary debates, testing the relation between the resulted metric of influence and speakers’ positions in government. The last case study builds a network of semantic and ideological affinity across UN General Assembly members, showing how graph-based methods can detect global political change. The thesis concludes with a discussion of potential future usages of the framework, as well as ameliorations.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.756068  DOI: Not available
Share: