309414 (v.1) Pattern Recognition 603


Area: Department of Computing
Credits: 25.0
Contact Hours: 3.0
 
** The tuition pattern below provides details of the types of classes and their duration. This is to be used as a guide only. For more precise information please check your unit outline. **
 
Workshop: 1 x 3 Hours Weekly
 
Syllabus: Pattern recognition is a field of applied mathematics which involves the description and analysis of measurements taken from a range of sources (e.g. sensor reading, financial data, psycho-physical measurements). In order to provide an effective, efficient description of patterns, pre-processing is often required to remove noise and measurement redundancy. Then a set of characteristic measurements ( features) and relations among these measurements are extracted for the representation of a specific pattern or model. Analysis (clustering, classification or identification) of the patterns with respect to a specific application is performed on the basis of this representation. The course will cover all essential aspects of the pattern recognition process. Specifically, the following will form the basis of the course - data acquisition, pre-processing, feature invariance, representation issues, feature selection and extraction, bayes classifier, clustering techniques, artificial neural networks (ANN), supportvector machines (SVM), hidden markov model (HMM) and Graphical Models.
 
** To ensure that the most up-to-date information about unit references, texts and outcomes appears, they will be provided in your unit outline prior to commencement. **
 
Field of Education: 20101 Formal Language Theory
Funding Cluster: 06 - Computing, Built Environment, Health
SOLT (Online) Definitions*: Supplemental
*Extent to which this unit or thesis utilises online information
Result Type: Grade/Mark

Availability

Availability Information has not been provided by the respective School or Area. Prospective students should contact the School or Area listed above for further information.