Ivo Düntsch http://www.cosc.brocku.ca/Faculty/Duentsch/
Department of Computer Science COSC4P76@cosc.brocku.ca
Brock University Room: J314, Phone: 3090

Machine Learning 4P76

Summary:
This half credit course provides an introduction to supervised learning and a variety of classification methods, as well as to concepts and procedures for evaluating and comparing the algorithms.

At the end of the course you should be able to
  • Explain the aims, history, and application fields of machine learning
  • Demonstrate an understanding of statistical and rule based methods and their differences
  • Discuss the advantages and disadvantages of various learning algorithms
  • Choose appropriate algorithms for a given data set and develop a setup for testing the prediction quality of each method
  • Validate results by using appropriate validation methods
Time & place: Tuesday, Thursday 15.30 - 17.00, venue TH248
Assessment: Two 90 min  open book class tests (20 % each), one group project (60 %)
Required reading: D. Michie, D.J. Spiegelhalter, C.C. Taylor (eds), Machine Learning, Neural and Statistical Classification.
This book is freely downloadable in PDF format.
Recommended reading: Margaret H. Durham, Data Mining

Tom Mitchell, Machine Learning

Tentative Lecture schedule
Week 1 Introduction to classification and machine learning
Week 2 Classical statistical methods
Week 3 Modern statistical methods I
Week 4 Modern Statistical methods II
Week 5 Revision - Class test 1
Week 6 Rules based learning I
Week 7 Rule based learning II
Week 8 Imputation and resampling I
Week 9 Imputation and resampling II
Week 10 Artificial neural networks
Week 11 Revision - Class Test 2
Week 12 Group presentations

Ivo Düntsch
September 9, 2009