SIT717 - Advanced Data Mining
|Enrolment modes:||(B, X)|
|Previously coded as:||SCC717|
|Assumed Knowledge:||Knowledge appropriate to the topic.|
|Unit chair:||G Li|
1 x 1 hour lecture per week, 1 x 2 hour practical per week
Note:Online teaching methods require internet access. Please refer to the most current computer specifications.
The unit will begin with an introduction to the advanced data mining processes such as CRISP-DM, then explain the requirements of business intelligence, in the context of customer relationship management. Methods to be taught in this unit includes variants of association rule discovery (for basket analysis); prediction techniques such as inductive inference of decision trees and Bayes models (for market prediction), clustering techniques such as self-organization maps (for market segmentation), but with emphasis on real world applications. A selection of recent real world business intelligence case studies will be incorporated in this unit to illustrate the introduced techniques. Advanced topics such as privacy-preserving data mining, text and sentiment mining will also be covered.
Projects (30%, 50%, 20%) 100%
Unit Fee Information
|Student Contribution Rate*||Student Contribution Rate**||Student Contribution Rate***||Fee rate - Domestic Students||Fee rate - International students|
* Student contribution rate for Commonwealth Supported students who commenced studies from 2010
** Student contribution rate for Commonwealth Supported students who commenced studies from 2009
*** Student contribution rate for Commonwealth Supported students who commenced studies from 2008
**** Fee rate for students who commenced studies from 2005
***** Fee rate for students who commenced studies from 2006
Please note: Unit fees listed do not apply to Deakin Prime students.