Computer Based Pattern Recognition

3 credit points

Syllabus, C-level, 1TT816

This course has been discontinued.

A revised version of the syllabus is available.

Code: 1TT816
Level: C
Subject(s): Computer Science
Grading system: Pass with distinction (5), Pass with credit (4), Pass (3), Fail (U)
Finalised: 24 May 2000
Responsible department: Department of Engineering Sciences

Entry requirements

Linear algebra. Multidimensional analysis. Numerical analysis. Mathematical statistics or a corresponding curriculum.

The course is intended to result in basic knowledge and understanding of how different types of learning

systems can be used for problems in computer based pattern recognition like linear and nonlinear

regression, classification, feature extraction, clustering, compression and visualisation. Applications are

taken from e.g. data mining and signal analysis.

Mathematical foundations of learning systems, basic concepts such as Bayesian statistical pattern

recognition,parametric and nonparameteric methods,discriminant functions, decision surfaces, parameter

estimation, supervised and unsupervised learning, generalisation, clustering. Families of methods like

decision trees, artificial neural networks, Bayesian learning, and prototype based learning. Applications

from e.g. data mining and signal analysis.

Lectures, lessons, seminares, laboratory work and homework.

Written and/or oral examination at the end of the course, homework assignments and laboratory work.