In the context of GIS there are many real world phenomena that can be represented as aggregrates of points. In many situations there is a need to be able to represent these aggregrates with a region.
These aggregrates, or dot patterns, and the regions, or footprints, that describe them have been widely researched (for a literature review see Dupenois and Galton 2009). However the existing work treats the patterns as static sets; in the real world they often represent objects that have tendencies to change, for example flocks and crowds. My research (supervised by Dr Antony Galton and Dr Jovisa Zunic) examines how the footprint of the pattern can be efficiently tracked as the underlying phenomena alters. Further to this we are looking at what information we can determine by recording the types of change undergone.
Along with Kent McClymont and David Walker, I have been looking at ways of classifying optimisation problems to see if they can be delineated in a way that shows which methods would be best suited to solve them. This combines work on heuristics, visualisation and dynamic data analysis. As part of this work we have also performed some data mining on the existing literature to identify whether or not there are clusters of researchers insulated from each other.
I am also currently engaged in a joint project with the Psychology Department at the University of Exeter to see if a game can be created that uses emotionally charged words to alter the emotional state of the player.