Peter Eades from the University of Sydney spoke on this topic at a dual-badged seminar between EVIMS2 and the CSIT department at ANU.
A graph consists of nodes and edges, a rather specific definition which threw me form the beginning. But Peter was able to start with a triangle, with Minard’s graph as an example of communication; John Snows’s graph as an example of discovery; and pretty pictures of Internet connectivity as an example of “gee whiz” that Peter says is a rather discredited way to advance knowledge. Now it’s all very well to talk about removing edge crossings for readability, but there are no axes to the graphs Peter discusses, and no requirement to show the point (1,2) one above (1,1). onetheless after some surprising observations and a nostalgic theorem, stress calculations, Jaccard similarity and multidimensional scaling all got a look in as the talk concluded.
John Rayner and Craig MacDonald gave this seminar on November 14. Craig started by claiming that most research is about creating, testing and modifying models. A conceptual framework is a system of concepts, assumptions, beliefs and theories that support and inform your research. In a science thesis, Craig says that generally the conceptual framework is given, so that your contribution is a modification to the conceptual framework. In many others, the framework is derived from literature. John described in much detail the results of the 18 in-depth interviews he conducted with recent honours and PhD graduates. Science typically came out with high scores, so while I might have been struggling to write the “conceptual framework ” section of a grant application, I suspect that like Moliere’s Bourgeois Gentilhomme, I’ve been using a conceptual framework all the time without knowing it!
I was honoured to be invited to attend a mid-project workshop run by the project team for IMSITE. All sorts of innovative projects are being implemented at universities from Townsville to Tasmania. This workshop served as a chance for the majority of the team to meet and present progress on their aspects of the project. other participants included school teachers, NSW Board of Studies staff, and university academics from nearby institutions. We were able to comment on the effectiveness of the projects to date, and also put up our hands to participate in further roll-out of strategies into new partner institutions in years to come.
I was particularly excited by the inquiry-based science instruction (inquiry-based statistics instruction, anyone?) and the concept of a single unit in Education for Science students to have a taste of what it’s like to study Education. I can only imagine how much better a tutor I would have been if I’d done a unit like that when I was an undergraduate.
A Hunter Valley based poster competition for schools, feeding into the The International Statistical Literacy Project, is also part of IMSITE. I hope that the new competition can dovetail with the Australian Statistics Competition in the future, so that maximum value can be gained for students.
The Canberra Branch of the Statistical Society of Australia held its Knibbs lecture on 4 November, and just before the talk I was presented with my Service Award certificate. The announcement of the award and opportunity to shake the hand of the Society’s National president, occurred at the Australian Statistics Conference in Sydney in July (see an earlier post). Here I am receiving the award from the Canberra Branch president, Ray Lindsay.
The 2014 Knibbs Lecturer for the Canberra Branch of the Statistical Society of Australia was Professor Noel Cressie, from the Centre for Environmental Informatics and National Institute for Applied Statistics Research Australia (NIASRA) at the University of Wollongong. His talk was on 4 November. Noel also received the Pitman Medal from the Society at the Society’s conference in July.
Noel’s talk pressed a number of the hot topics in statistics at the moment – big data, spatial methods, non-Gaussian data, it was all there. Big data for him meant 2.7 million rows of data, from remote sensing equipment aiming to predict cloud cover. The spatial component of this is pretty obvious, and the non-Gaussian bit comes from the binary nature of the predictions required (is it cloudy or not? And note that a 5% chance of cloud is defined as a cloud!) MODIS is the name of the gadget that collects the data Noel and his colleagues analyse, and the change-of-support is to do with the degree of granularity at which prediction is required – a grid of 0.5 degree squared, or maybe five km squared, or even one km squared.
Noel is a great presenter, I really enjoyed his talk.
The two discussants were Philip Kokic, of CSIRO; and Daniel Elazar, of the Australian Bureau of Statistics. As a Knibbs discussant in a previous year I know this is a delicate balancing act. While you do get a copy of the talk in advance, you still have to make your comments both reflective of your understanding of the lecture, while still having the opportunity to talk about your own work in the are of the main talk. I think they both achieved that balance.
Mark Luu presented his initial PhD seminar on this topic on 31 October too. His abstract is as follows. “This research project involves investigating novel approaches for analysing free-form unstructured text data in the medical domain (particularly in the Nursing Handover Clinical Context). The proposed research on intelligent information extraction will draw upon novel text mining, natural language processing (NLP) and Probabilistic graphical modelling techniques for processing unstructured text data, leading to extraction of structured information, which can then be better utilised for improving handover processes, for enhancing patient safety and quality of care.”
The better use of text to communicate in medicine is something that has interested me for a number of years. I think if he is able to access appropriate outcome data to see if his method really works, then this also can be a valuable contribution.
Robert Phan presented his introductory PhD seminar on this topic on 31 October. his abstract is as follows. “This research project involves investigation into large-scale machine learning techniques for better understanding of complex health data – with particular focus on analysing clinical notes for adverse event monitoring, decision support alerts and automatic clinical summaries. The proposed research will draw upon novel text mining, natural language processing (NLP) and machine learning techniques for processing unstructured text data, leading to extraction of structured information, which can then be better utilised for improving clinical handover processes, with an objective to enhancing patient safety and quality of care.”
I think his main contribution will be in hybrid methods, combining several existing tools to arrive at a stronger result.
That’s a very broad brief, set by the London Workshop on the future of the statistical sciences (which is equally broad). There’s a great set of seven modern areas where statistics is making a contribution, which should capture the attention of the public straight away. Then there’s a collection of big hairy problems – anyone keen to tackle big data, climate change, reproducibility of research? But the idea that I will try to use in my interactions with colleagues into the future is that statistics is not well descrbibed by the phrase “the science of data”, but much better by the phrase “the science of uncertainty.” That’s something that many scientists don’t necessarily have, the understanding of uncertainty, and that’s what I want to instil in my students.
You can find the report at www.worldofstatistics.org.
About 30 academics, PhD students and academic support staff gathered at ANU on 20 October to hear the words “from the horse’s mouth” about OLT grants for 2015. I do wonder about how willing an institution would be to let go of a successful program that might otherwise be thought of as a point of difference. But that aside, the session on impact was really valuable and I came away with a couple of ideas bouncing around in my head.
Well the Brits know what they want and they have a few ideas as to how to get it! This is an inspiring document setting out where the UK wants to be in 2030 in terms of STEM education. It’ll be interesting to see how they get on in the next decades.