My PhD work focused on spatio-temporal data mining. Data containing temporal and/or spatial features present further complications to the analysis of already high-dimensional data sets. Due to the ever increasing ways in which data can be collected, spatio-temporal data bases are becoming more commonplace especially in areas such as bio-medical data analysis and robotics.

The advent of Proteomics has seen an explosion in post-translational and post-experimental data sets. This provides an analysis problem for experts working in this field. My current research work centers on the use of Artificial Intelligence techniques to perform data mining upon, amongst others, Proteomic data sets, in particular in the area of 2-D Electrophoresis Gel post-experimentation data and microarray expression data.

Such important knowledge will help to speed up this time consuming and complex analysis process. It is hoped that the detection of patterns and trends within such data may eventually help in the early detection of diseases; of particular importance when early treatment of a disease results in good recovery results (e.g. Cancers), and assist drug development by providing novel drug targets.

Research in collaboration with Dr Ken McGarry and Professor Stefan Wermter's Hybrid Intelligent Systems group performed an analysis of spatio-temporal robot behavioral data. Providing transparency to the underlying actions performed by the robots used in these experiments can aid understanding and allow the identification of important attributes, key time points and reusable knowledge in the form of rules. This can be of particular interest in the area of imitation learning, in which a 'teacher' robot performs an action and a 'student' robot attempts to mimic this through observation.