Skip to main navigationSkip to main content
The University of Southampton
Centre for Risk Research

Seminar highlights how to find the hidden meaning in data

Published: 17 March 2016
Data = Shape = Meaning

In a recent Centre for Risk Research seminar, Conrad D'Souza introduced his work applying novel methods in Topological Data Analysis (TDA) to the study of competitive events.

Conrad explained that TDA is founded on the belief that data has shape - and this shape has meaning. He explained an approach for giving shape to competitive event data - specifically horse racing - and that by understanding this data as a simplicial complex and bringing theories from topology to bear, this complex can be analysed and a signal measuring the underlying quality of the competing alternatives found.

His results indicate that the speculative markets surrounding horse racing do not fully account for the information contained within past performance indicators. However, by applying TDA, he has found that more information can be extracted, enabling more accurate predictions to be made.

Conrad D'Souza is a Web Science PhD student working jointly with Mathematics and the Centre for Risk Research, University of Southampton. His research aims to develop new analytical tools for exploring data by studying the shape of data and demonstrate the economic value of understanding data through this paradigm.

Privacy Settings