Date and time: 3:00pm-4:30pm, Wednesday 18th October, 2006
Venue: 300.217
Abstract: This talk is a brief introduction to the family of tree-based regression methods. Tree-based methods are very useful modelling tools which not only give a good prediction, but also a good representation of the relative importance of each decision variable. We will demonstrate methods such as regression tree, bagging (Boostrap AGGregaTING), random forests, boosting and applying them to data from Australian wool auction. Alternate methods such as neural network and their link with tree methods will also be discussed.
About the speaker: Kyle Chow is a PhD candidate in the Department of Mathematics and Statistics, Curtin University of Technology.
Seminars are free and open to all postgraduate students in the Department of Computing and the Department of Mathematics and Statistics. No booking is necessary and biscuits, tea and coffee are provided. Seminars are as informal as the speaker desires. If you are not a postgraduate student and wish to attend a seminar you may only do so with the express permission of the relevant speaker(s).
If you are interested in giving a presentation in this seminar series, or to make suggestions for speakers, please contact Simon Puglisi (Computing): or Christina Burt (Maths):
the seminar co-ordinators. Seminars are normally held at 300.217 (building 300, level 2, room 217) - you can find where this is exactly using the Bentley campus map.
