School of Computer Science
University of Nottingham
Nottingham, NG8 1BB UK
Email: jxz@cs.nott.ac.uk
Office: C50
My research interests include the modelling of complex social interactions using agent-based technologies.
I am currently working towards a PhD; the working title of my thesis is Agent-based Models of Group Decision Making.
This work is supervised by Dr. Brian Logan and Dr. Natasha Alechina and is carried out as part of the activities of the Agents Lab at the School of Computer Science, University of Nottingham.
In semester one of 09/10, together with Tim Brailsford I am teaching module G52IWS - Internet and Web Services.
In semester two of 08/09 I was teaching assistant for module G54DIA - Designing Intelligent Agents. I also maintain the supplementary material for this module.
In semester one of 08/09 I taught module G52IWS - Internet and Web Services.
Prior to commencing PhD studies I completed an MSc in Management of Information Technology; the title of my dissertation was Multi-agent Simulation of Group Decision Making in Animals.
Abstract
The application of multi-agent simulation to the study of group decision making in animals is explored through the creation of a software application. The software is applied to the replication of a recent simulation study, conducted by Sellers, Hill & Logan (2007), concerning group decision making in a troop of chacma baboons (Papio hamadryas ursinus). The importance of replication as a scientific tool is investigated, permitting study of the techniques, merits and pitfalls of this approach. Comparisons of data from both the original simulation study and direct field observations are made as incremental improvements to the quality and predictions of the model are achieved. The findings of this work both support and extend findings of the original study. Increasing the realism of the environmental model is shown to have a noticeable effect on the group decision making process, providing initial evidence that the ways in which group decisions are made can be influenced by environmental conditions.
A short presentation regarding this work is available, you might also be interested in the full text.
The software developed during the course of this work requires Sun's Java and the MASON Multiagent Simulation Toolkit.