Research Theme: Challenging data

Modern data comes in all sorts of different forms that do not necessarily fit well with traditional data analysis methods. For example, it can occur in the form of images, text, visual displays or mathematical functions. The size of datasets can be much larger than those traditionally analysed and the speed at which decisions have to be made about such data can be much faster than previously required. Under this theme we explore these data sources different types of challenging data and develop new methods to explore such data for its analysis.

Research Theme: Challenging data Projects

Anomaly detection in noisy time series

The ability to detect anomalies or outliers in time series, and to develop algorithms that can find them at the earliest possible sign of a deviation from expected behaviour, is critical for many applications including those arising in security, epidemiology, and monitoring of critical infrastructure. This project aims to develop new methods to characterise the expected time series behaviour of systems, even in the presence of significant noise and concept drift, and with spatio-temporal and multivariate time series.

Bayesian Approaches to Modelling and Analysis of Big Data

Bayesian methods for modelling and analysis of data are now well established. However, as with many statistical methods, their applicability in the context ‘big data’ is still being explored. This project will consider four questions that relate to ‘scalable Bayes’, that is, Bayesian models and computational methods that scale up to big data problems.

This project links to a number of other projects across the Centre.

From Cells to Organs: Exploring Variability in Physiological Systems

Everybody is different, and every body is different. Significant variability is a common feature of all of the physiological systems that compose the function of the human body, and understanding this variability is critical to explaining differences in susceptibility to pathological conditions, and also to explaining how medical treatments can potentially succeed or fail even when applied to treat the same condition.

Improving returns from Southern Pine plantations through innovative resource characterisation - virtual log models

Investigation and development of virtual log models for Southern Pines will be based on analysis of data from the cores, peeled billets and approximately 60 sawn logs. We plan to predict log and stem wood properties from the breast height cores taken in the field study. Following this, applied mathematics will be used to investigate the processing of these virtual logs and predict properties for the virtual boards ‘sawn’ from these logs.