You are responsible for coming prepared to class. This includes reading through the material before attending class. You will get a lot more out of the lectures and discussions in this manner. It is cliché, but true. Each week will follow a similar pattern. The course focuses both on the theory of data mining, machine learning, and its practical application. Most classes will start with an exam over the previous week’s material. followed by a lecture on new material. Followed by an interactive exercise focusing on the theory. Followed by an interactive lecture/exercise focusing on the application. This will sometimes be guided. sometimes in groups, and sometimes individually. The schedule below is tentative and subject to change. You must check it regularly.
The following are a list of topics and the order in which we will approach them. The finalized schedule will be posted each week to OAKS.
Syllabus discussion. Introduction to data mining and machine learning and Bayesian classification.
Week 2 and 3:
Neural Networks and Linear Discriminants (Chapter 3)
Week 4 and 5:
Multi-layer Perceptron (Chapter 4)
Week 6 and 7:
Dimensionality Reduction (Chapter 6)
Week 8 and 9:
Support Vector Machine (SVM) (Chapter 7)
Week 10 and 11:
Decision Trees (Chapter 12) and Ensemble Methods (Chapter 13)
Week 12 and 13
Unsupervised learning (Chapter 14)
Week 14 and 15
Last week of classes!