ANU Data Mining Group

News:

• 16 October: Added a list with the upcoming student presentations.

• 13 October: Added the slides for lectures End-to-end data mining (Graham) and Privacy-preserving data mining (Peter).

• 4 October: Added the assignment 4 (PDF).

• 3 October: Added the slides and script for Markus' lectures.

• 26 September: Added the assignment 3 (PDF).

• 25 September: Added the updated course schedule.

• 26 August: Added a link to the 2nd assignment.

• 25 August: Added slides for Graham's last two lectures (see below).

• 24 August: Added slides for Graham's 4th and 5th lectures (see below).

• 22 August: Added slides for Graham's 3rd lecture and documentation on Rattle (see below).

• 18 August: Added slides for two of Graham's lectures (see below).

• 15 August: Added the updated course schedule.

• 25 July: Added the course schedule and assignment for module 1 (PDF).

• 20 July: Added a link to John's Web site with material for labs (see bottom of this page).

• 19 July: Updated lecture times:
  • Lecture A: Tuesday, 11-13 in room G35 (John Dedman building, see details below).
  • Lecture B: Thursday 10-11, in room G35.

  Slides for the first five lectures are now available (see below).

• 17 July: The lecture and lab times and venues are:
  • Lecture A: Tuesday, 11-12 in room G35 (John Dedman building, see details below).
  • Labs: Tuesday 2-4, Coombs computing laboratory.
  • Lecture B: Thursday 10-11, in room G35.
  • Lecture C: Thursday 13-14, in room G35.

• 30 June: Initial version of the course web site made public.

Lecture and lab times and venues

• Course schedule (note: slight changes might still happen).

• All lectures are being held in the John Dedman Mathematical Sciences Building [no. 27], room G35.

  Lecture times:

  1. Tuesday 11-13
  2. Thursday 10-11

  Please note that not all lecture slots will be used in all weeks.

• Labs are Tuesday 2-4, in the Coombs computing laboratory.

Course coordinator

John Maindonald

Prerequisits
We don't actually have formal prerequisites. We assume that participants have basic computing, statistical and mathematical skills with honours level expertise in either of these areas. Our intended audience are honours and graduate students in these areas. We intend to look at both mathematical (algorithm) and computing issues.

In the course we will analyse algorithms, do some coding and look at efficiency issues, and also do some hands on data analysis (data mining).

Lecturers

• Peter Christen
  ANU Department of Computer Science, FEIT

• Graham Williams
  Togaware / ATO / ANU Department of Computer Science, FEIT

• John Maindonald
  Centre for Mathematics and its Applications, MSI

• Markus Hegland
  Centre for Mathematics and its Applications, MSI

• Stephen Roberts
  Department of Mathematics, MSI

• Alan Welsh
  Centre for Mathematics and its Applications, MSI

Course Assessments

The assessment for this year's course is as follows:

1. Four assignments, two worth 10% and two worth 15%.
2. Paper presentation worth 20% (in weeks 12 and 13).
3. Examination (take home or written) worth 20%.
4. The final course mark will be the sum of the assignments, presentation and exam marks.

Course Material

• Welcome and introduction
  (Peter Christen)
  Lecture slides (4-up): Lect-0, PDF

• Data mining overview, process, data issues, data quality, data preprocessing, data integration / linkage
  (Peter Christen)
  Lecture slides (4-up): Lect-1, PDF     Lect-2, PDF     Lect-3, PDF     Lect-4, PDF

• Lecture material for John's modules is available from John's MATH3346 lab Web site.

• Data mining algorithms
  (Graham Williams)
  Lecture slides: Introduction, PDF, 6-up     Cluster Analysis, PDF, 6-up     Association Analysis, PDF, 6-up     Decision Trees, PDF, 6-up     Ensemble Decision Trees, PDF, 6-up     Support vector machines, PDF, 6-up     Neural Networks, PDF, 6-up    

• Algorithms for association rules / MARS
  (Markus Hegland)
  Lecture slides: slides
  Script: script

  Markus' lectures are based on:

  • Data Mining - Challenges, Models, Methods and Algorithms
    Draft, February 2003.
    Paper (pdf, 884 KB)

    and

  • Algorithms for Association Rules
    Tutorial in the Program Mathematics and Computation in Imaging Science and Information Processing of the Instute for Mathematical Sciences, NUS, Singapore 10.-12. December 2003.
    Slides (pdf, 323 KB)

• End-to-end data mining (Graham): Slides (PDF, 6-up)

• Privacy-preserving data mining (Peter): Slides (PDF, 8-up)

Lab material is available from John's MATH3346 lab Web site.

Documentation on Rattle (Graham's lab): rattle.pdf.

Assignments

• The assignment for module 1 (PDF) is due on Friday 11 August, 5 pm.

• The 2nd assignment (Graham Williams) is due Friday 22 September 2006.

• The 3rd assignment (John Maindonald) is due Tuesday 10 October 2006.

• The 4th assignment (John Maindonald) is due Friday 20 October 2006.