Programme
9:15-9:45 Arrival and coffee
9:45-10:00 Welcome – Alex A. Freitas (University of Kent)
10:00-11:15 Session 1, Chair: George Smith (University of East Anglia)
Frans Coenen - Association Rule Mining in the Wider Context of Text, Images and Graphs (download slides of this talk)
Jenny Harding - Data Mining Applications in Manufacturing (download slides of this talk)
11:15-11:30 Coffee
11:30-12:45 Session 2, Chair: Colin Johnson (University of Kent)
Andy Secker - An Experimental Comparison of Classification Algorithms for the Hierarchical Prediction of Protein Function (download slides of this talk)
Niall Rooney - Stacking for Supervised Learning (download slides of this talk)
12:45-13:45 Lunch
13:45-15:00 Session 3, Chair: Beatriz de la Iglesia (University of East Anglia)
David Watkins - Analysis of the Effect of Sample Size on the Quality of Data Mining Models (download slides of this talk)
James Cussens - Bayesian Classification and Regression Trees (download slides of this talk)
15:00-15:15 Tea
15:15-16:30 Session 4, Chair: Frans Coenen (University of Liverpool)
Beatriz de la Iglesia - Application of Multi-Objective Metaheuristic Algorithms in Data Mining (download slides of this talk)
Discussion and End
Biographies
Frans Coenens, University of Liverpool
Frans Coenen is a senior lecturer within the Department of Computer Science at the University of Liverpool. He has many years of experience within the field of KDD working with both academic and commercial partners and has published widely in the area. He initiated the original concept of the UKKDD series of symposia. He has been and is currently on the programme committee of a large number of KDD and related conferences. His current work is focussed on Multi-Agent Data Mining (MADM), and text and image mining.
James Cussens, University of York
James Cussens (JC) has a Maths BSc from Warwick and a Philosophy PhD from King's College London. After working on a number of machine learning research projects in London, Glasgow and Oxford he took up a lectureship in York in 1997 and remains there to this day. JC works on machine learning with particular interests in: inductive logic programming, natural language and Bayesian approaches. Together with Nicos Angelopoulos he developed the MCMCMS system which uses stochastic logic programs to define priors on the structures of statistical models and uses MCMC to sample from posterior distributions. JC is an Associate Editor of the Journal of Artificial Intelligence Research and is on the editorial board of Machine Learning journal. He has recently joined the York Centre for Complex Systems Analysis.
Beatriz de la Iglesia, University of East Anglia
Dr Beatriz de la Iglesia, PhD MBCS was born in Asturias, Spain in 1967. She gained a First Class BSc in computing sciences at the University of East Anglia (UEA) in Norwich, 1994. She gained her PhD in 2001 investigating the development of meta-heuristic algorithms for data mining at the same University. She became a lecturer at the School of Computing Sciences, UEA in 2002. Her main research interest at present are the investigation of multi-objective metaheuristics for data mining (funded by EPSRC); the investigation of text mining algorithms, in particular with application to medical data, and the more general theme of medical data mining. She has published over 30 papers related to data mining and optimisation.
Jenny Harding, Loughborough University
Dr Jenny Harding is a Senior Lecturer in the Department of Mechanical and Manufacturing Engineering at Loughborough University. She has substantial industrial experience having worked for over 15 years in the engineering and textile industries before joining Loughborough University in January 1992. Her expertise includes knowledge management and reuse, tools to support knowledge sharing within collaborative teams, knowledge discovery and data mining applications in Manufacturing, and 'Best Practice' information and knowledge. Her research has been funded by Europe and in the UK by EPSRC and Industry. She has a wide range of academic publications and has supervised several successful PhDs in these subject areas.
Niall Rooney, University of Ulster
Niall Rooney has a BSc. in Information Technology from Queens University, Belfast (1992) and a MSc. in Computation from Oxford University (1996). Niall had worked a number of years in the software industry prior to becoming a researcher in the Northern Ireland Knowledge Engineering Laboratory (www.nikel.ulster.ac.uk) at the University of Ulster in 2000. His primary research interests include machine learning, case-based reasoning, and information retrieval and he has published 20 refereed papers in these fields. He has acted as reviewer for the following journals: IEEE Trans. PAMI, IEEE Trans. TITB and Information Processing and Management. He is also in his final year as a part-time PhD student in the area of ensemble meta-learning techniques for regression.
Andy Secker, University of Kent
Dr Andrew Secker received a first class B.Sc. with honors in Computer Science from the University of Kent, UK, in 2002 and was awarded a Ph.D. from the same institution in 2006. Andy's Ph.D work centred on the application of immune metaphors to the filtering and retrieval of interesting information from the web. At present, Andy is working as a Research Associate for the Computing Laboratory at the University of Kent and is currently investigating the application of artificial immune systems to the prediction of GPCR (G-protein coupled receptor) protein function.
David Watkins, Solution Architect for SPSS
David Watkins is a Solution Architect for Predictive Analytics at SPSS. He works with SPSS’ clients to design and deliver predictive analytics solutions to their business problems across industries including banking, finance, insurance, telecoms and retail. David has delivered many projects involving the management of predictive analytics, including systems to automatically refresh and score thousands of predictive models, all utilizing the CRISP-DM methodology. He also sits on the consortium for CRISP-DM. Previously he was one of the founding developers of the Clementine data mining system, which included development of the automated neural network engine.