DE Lunch (Multidisciplinary Research Week Event)
Date: 10.02.2012
Speaker: Lorraine Warren and Luke Greenacre
Location: B85/2209
The next Digital Economy Lunch will feature presentations by Lorraine Warren and Luke Greenacre
Lorraine Warren’s recent research has focused on the complex dynamics of innovation systems, particularly early stage concept development, the latter developed through evaluation of best practice in the US, in particular North Carolina State University and the University of San Diego. She has advised Russian technology institutes on issues of technology transfer, and explored these issues within the Connecting Communities for the Digital Economy initiative (EPSRC). Lorraine is on the Steering Group of the University’s Web Science Doctoral Training Centre and a member of the University’s Digital Economy Research Group. She heads the INSPIRE Strategic Partnership with IMSciences in Pakistan.
For an introduction to Lorraine’s work please see: http://www.edshare.soton.ac.uk/5881/1/lorraine_warren.html
Temporality, emergence and value in the creative industries: questions of theory and methodology
The rapid development of digital technologies and the growing availability of digital data and digital content has enabled new forms of communication, connectivity and international influence. The emergence of technologies from the hands of computer scientists into public domains (as the ‘Digital Economy’) has enabled the possibility of resetting the relationship between the individual and the corporation, the specialist and the everyday user. It should now be easier than it has ever been to not only access and use new technologies, but to extend them, customise them, develop new combinations, to improve, radically innovate and disrupt how we live our lives and create new value, new futures. However the roadmap for inductive thinking that will create value in novel and unforeseen ways in new contexts and settings is not clear; classical models of innovation may be too focused on economic value creation at the expense of wider societal benefit. This talk will address the methodological challenges of researching interdisciplinary, multifaceted ephemeral phenomena.
Luke Greenacre started with the University of Southampton in 2011. He has previously worked at the Centre for the Study of Choice and the Marketing Discipline Group at the University of Technology in Sydney; the University of Sydney; the University of Western Sydney; and the Australian Institute of Music. His PhD examined how to model sender communication decisions using discrete choice experiments. His research since then has considered modelling decision variance, capturing peer behaviour in social networks, and education practice.
Luke’s presentation will outline the theoretical and methodological basis for a new approach to predicting online referral networks. Through the integration of discrete choice experiments and agent based modelling techniques it is intended to build models of how individual agents actually behave in real markets, and then observe how this behaviour aggregates at the network level. It is intended that this will allow us to understand how small changes in communication decisions by individual actors can alter market outcomes.
A five stage modelling procedure will be outlined. The first four stages are aimed at generating stable individual level models of how actors a behaving regarding their online referral behaviour. These four elements being modelled in these stages include the present utility function underlying consumers’ present product choices; the choice of sender by those seeking referral information; the choices of senders regarding to whom they will provide referrals; and the effect of various referrals on consumers’ utility functions. Discrete choice experiments allow each of these can be modelled such that choices can be mapped as a function of the features of the choice alternatives within each element. With these functions identified they can be included as inputs into an agent based model to identify the simultaneous solution for each of these individual behaviours at the network level.
The purpose of this presentation is to discussion the potential pitfalls of this type of methodological approach, and how this form of multi-disciplinary/multi-method approach may also yield substantial advancements in present models of this type of communication behaviour.