Monash Data Mining Centre
The Monash Data Mining Centre (MDMC) is a research and consulting group with a strong international reputation, located within the Faculty of Information Technology at Monash University. The Centre contains academic and research staff and research students.
We provide advanced research and training in data mining for business, government, science and industry. In particular, we offer:
An organisation considering data mining often faces many decisions regarding how data mining can help, which software products they should purchase, whether their existing staff should be trained in data mining, or whether they recruit data mining personnel, etc. A collaboration with MDMC typically involves researchers using a sample of the organisation's data to demonstrate the potential of data mining, and providing recommendations about how the organisation should proceed. We then train the relevant staff and/or recommend graduates. Or, for single projects, we can quote our price for doing some or all of the analysis for you. Finally, we can identify relevant research issues of mutual interest for future collaborations.
Members of MDMC have experience in many areas of data mining, and many different tools including both in-house specialties and enterprise-wide commercial packages. Follow the Applications link to see descriptions of our major areas of data mining work. Follow the Facilities link for a survey of packages, resources, and techniques we use. Follow the Software link to see descriptions of in-house programs that lead the world in knowledge discovery. (Most of our software is freely downloadable under an academic license, and other licenses are available.)
EducationAcademic staff in MDMC offer a number of courses to industry:
Of course we also teach many units in undergraduate and postgraduate programmes related to data mining including:
These can be taken as part of a degree or as stand-alone units.
Minimum Message Length Induction (MML)MML was developed by Chris Wallace of Monash in 1968. It has been widely and successfully applied, in such areas as machine learning, artificial intelligence, statistics, econometrics and data mining. It has all the advantages of Bayesian methods (such as learning from small samples), plus invariance to scale transformations. Software based on MML routinely outperforms other algorithms. Furthermore, MML can be used to compare models found by non-MML means. For more information see Lloyd Allison's MML web page and/or David Dowe's MML web page. A major part of MDMC research involves extending the foundations of MML, and applying it to different kinds of models, such as neural nets.
Contacts and Links
To obtain further information, please contact the MDMC Centre
Last Updated: Wednesday 11 September 2002 15:13:10