Pattern Recognition

Lecturer: John Quinn
Teaching takes place Thurs 1800-2100 in new CIT building, small lecture theatre 5.

This short course is a mixture of theory and practice in pattern recognition. Programming exercises take place regularly throughout the semester, using Python (with useful libraries SciPy, NumPy, and Matplotlib).

Lecture notes are available below, but they only contain an overview and supporting material for the classes—attendance is essential.

Class material

Week 1 notes: introduction to pattern recognition.
Week 2 notes: distance functions and neighbours.
Practical exercise 1: MNIST digit classification.
Week 4 notes: principal components.
Practical exercise 2: Ugandan eigenfaces.
Week 6 notes: decision trees.
Practical assignment: classification of breast cancer.
Week 7 material for class exercises: naïve Bayes.
Practical exercise 3: logistic regression to predict patient mortality.
Week 9 notes: clustering and k-means.

Page maintained by John Quinn