Skip to content

News

Published on

It is a great honour to receive the IEEE ICDM 2023 10-year Highest-Impact Paper Award for "Dynamic Time Warping Averaging of Time Series allows Faster and more Accurate Classification" by François Petitjean, Germain Forestier, Geoff Webb, Ann Nicholson, Yanping Chen, and Eamonn Keogh.

This was my first paper in the field of time series classification, and consequently my technical contribution was relatively minor, further evidence, in case anyone needs it, that it pays to collaborate with amazing researchers!

I had no idea at the time that this would in time grow to be my primary field of research!

Published on

Angus Dempster has received the Computing Research and Education Association of Australasia Distinguished Dissertation Award, the highest Australasian recognition of a computer science PhD for his exceptional PhD that introduced the revolutionary ROCKET approach to time series classification.

He gives an overview of this work in the following video.

Published on

I am honoured to have been recognised in The Australian's 2023 Research Magazine as Australia's top Bioinformatics and Computational Biology researcher. This is the third year in a row that The Australian has recognised my research in this way. This is a great recognition of the fantastic research of my bioinformatics collaborators, most especially Jiangning Song. This year they also recognised me as Australia's leading Data Mining & Analysis researcher.

Published on

I am honoured to have been recognised in The Australian's 2023 Research Magazine as Australia's top Bioinformatics and Computational Biology researcher. This is the third year in a row that The Australian has recognised my research in this way. This is a great recognition of the fantastic research of my bioinformatics collaborators, most especially Jiangning Song.

Published on

We introduce a novel incremental decision tree learning algorithm, Hoeffding Anytime Tree, that is statistically more efficient than the current state-of-the-art, Hoeffding Tree. We demonstrate that an implementation of Hoeffding Anytime Tree–-"Extremely Fast Decision Tree", a minor modification to the MOA implementation of Hoeffding Tree–-obtains significantly superior prequential accuracy on most of the largest classification datasets from the UCI repository. Hoeffding Anytime Tree produces the asymptotic batch tree in the limit, is naturally resilient to concept drift, and can be used as a higher accuracy replacement for Hoeffding Tree in most scenarios, at a small additional computational cost.

Download the paper here.

Download the software here.