FIT5169 - Probabilistic expert systems

Leader(s): Kevin Korb

Offered

Not offered in 2009

Synopsis

This unit provides an understanding of current methods of automated probabilistic reasoning in graphical models and their application in building expert systems. Techniques for data mining graphical models will also be surveyed. A theoretical background in deterministic and stochastic probability propagation in Bayesian networks is joined with a case study of application development in a domain such as ecological risk assessment or meteorological modeling.

Objectives

At the completion of this unit students will:

1. be able to design and build probabilistic expert systems;
2. understand the application of probability theory to reasoning under uncertainty;
3. be able to apply automated decision analysis tools;
4. be familiar with the main Bayesian network tools and their capabilities;
5. understand how to data mine graphical models;
6. be able to knowledge engineer Bayesian networks.

Assessment

Bayesian network modeling: 25%; Data mining Bayesian networks:25 %;
Exam, Department-Open book (3 hours): 50 %.

Contact hours

2 hours of lectures/week; 1 hour of tutorials/week

Prerequisites

For MAIT students, FIT9017, FIT9018, FIT9019, FIT9030, FIT9020 and FIT4037.

Additional information on this unit is available from the faculty at:

http://www.infotech.monash.edu.au/units/fit5169