sbutler.com

Research Publications

Here is some of my work:

• Butler S. M., G. I. Webb (2005). Mining Group Differences. In Wang, J. (ed) Encyclopedia of Data Warehousing and Mining. Idea Group Reference. [ISBN: 1-59140-557-2]
• K. K.W. Siu, S.M. Butler, T. Beveridge, J.E. Gillam, C.J. Hall, A. H. Kaye, R.A. Lewis, K. Mannan, G. McLoughlin, S. Pearson, A.R. Round, E. Schultke, G.I. Webb and S. Wilkinson (2005). Identifying markers of pathology in SAXS data of malignant tissues of the brain. In Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 548, 140-146.
• Butler S. M., G. I. Webb and R. A. Lewis (2003). A Case Study in Feature Invention for Breast Cancer Diagnosis Using X-Ray Scatter Images. In Proceedings of the 16th Australian Joint Conference on Artificial Intelligence (as 'Lecture Notes in Artificial Intelligence').
• Webb G. I., S. M. Butler and D. Newlands (2003). On detecting differences between groups. In Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 256-265. ACM Press. Only 34 papers were accepted out of 258 submitted. Alternate download here.
• Butler, S. M. (2002). An investigation into the relative abilities of three alternative data mining methods to derive information of business value from retail store-based transaction data, Honours Thesis, Deakin University.

Research History

Masters Research

March 2003 - June 2006

I was a student at Monash University's School of Computer Science and Software Engineering (CSSE) in Australia. I was supervised by Prof. Geoff Webb, Prof. Rob Lewis and Dr. Karen Siu.

My research focused on using data mining to classify breast tissue as either normal (healthy tissue) or tumorous using X-ray diffraction data. This work is in co-operation with members of the School of Physics and Monash Centre for Synchrotron Science. I am also involved with the Victorian Bioinformatics Consortium (VBC) which is based here at Monash University and brings together researchers from academia, industry and government.

Honours Research

Final year studies (2002)

The major component of my final honours year was a research project. Many retailers collect large amounts of data, so my honours research, which was conducted under the joint supervision of Prof. Geoff Webb and Dr. Douglas Newlands, focused on application of several data mining techniques within the retail industry.
First, three data mining tools were selected: C4.5rules (decision trees), Magnum Opus (association rules) and STUCCO (contrast sets). Then experiments were run on retail transaction data. The outputs of the systems were compared and contrasted and domain experts gave input as to the understandability, unexpectedness and potential value of each rule. This allowed for a comparison of each systems suitability to the retail domain.
Contrary to Bay and Pazzani's assertion that association rules were unsuitable for contrast-set mining, our experiments showed association rules were able to be used to find differences between groups. This lead to further work, which developed the theoretical proof and was then published at KDD2003.

Research Assistant

July 2000 - July 2001; March 2003 - Oct 2004

Between 2000 and 2001 I was a member of the Knowledge Acquisition, Processing, and Management Group at Deakin University, working under the guidance of Prof. Geoff Webb. During this time I worked to use, develop and maintain several Machine Learning algorithms, including Lazy Bayesian Rules and Naive Bayes.
I am now working in a similar position although this time from Monash University.