Zena Wood

We are surrounded by collective phenomena: traffic jams, flocks of birds and queues; even we could be considered as a collective (i.e., a collection of organs). We use the term "collective" to refer to groups of individuals or individual entities that are bound together by some principle of unity (the source of their collectivity) that makes it natural to consider them as collectively constituting a single entity. Currently, my focus lies on those collectives whose members can be assigned physical locations; for exampe, crowds, herds of animals and groups of biological cells would be considered as possible collectives but not the set of natural numbers.The ubiquity of collective phenomena and their movement patterns should mean that adequate tools exist that allow formal reasoning about these phenomena. However, this does not appear to be the case. My PhD entitled 'Detecting and Identifying Collective Phenomena within Movement Data' focused on the development of a method that would allow collective phenomena to be detected and identified within a spatiotemporal dataset. A prerequisite for this research was to understand the range of collectives that exist and the different movement patterns that they may exhibit.

To identify the different collectives that exist a range of example collective phenomena were examined. This examination resulted in five features being identified as possible classification criteria: membership, location, coherence, roles and depth. Full details of this taxonomy can be found in 'A Taxonomy of Collectives' (Wood and Galton 2009).

A method has been proposed that allows the presence of spatial collectives to be identified within a spatiotemporal dataset. Full details of this method can be found in Wood and Galton 2011 and Wood 2011. Results of the application of this method can be found here.

The movement patterns that are exhibited by collectives are probably one of the most imortant features of a collective that we may wish to reason about. However, it would appear that we do not currently possess the necessary tools. I am currently developing a framework that allows collective motion to be modelled and reasoned about. The framework, which allows motion to be viewed on multiple levels of granularity, takes into account the importance of the relationship between the motion of the collective and those of its individual members. I am writing an initial computer program to establish the efficiency of the framework. Running data has been collected from various races from which examples of movement patterns are hoped to be extracted by the program that can be adequately described using the framework. Details of this framework can be found in 'Zooming in on Collective Motion.