Compositional Bayesian modelling and its application to decision support in crime investigation

Qiang Shen, Mark Lee, Jeroen Keppens

Research output: Contribution to conferencePaperpeer-review

31 Downloads (Pure)

Abstract

Despite increasing interest in the development of intelligent techniques to aid in the prevention and detection of crime, an important issue that has not yet been addressed by existing work is the use of knowledge based systems (KBS) to aid in the synthesis and analysis of hypothetical scenarios in major crime investigation.The main limitation of conventional KBS approaches is their lack of robustness to deal with the substantial variability of crime scenarios. This paper introduces a method to apply model based reasoning techniques to this problem. In particular, the existing compositional Bayesian modelling approach is extended and adapted to create hypothetical crime scenarios. Also, methods developed in the area of model-based diagnosis are used to support the analysis of synthesised crime scenarios.
Original languageEnglish
Pages138-148
Number of pages11
Publication statusPublished - 2005
Event19th International Workshop on Qualitative Reasoning - Graz, Austria
Duration: 18 May 200520 May 2005

Conference

Conference19th International Workshop on Qualitative Reasoning
Country/TerritoryAustria
CityGraz
Period18 May 200520 May 2005

Fingerprint

Dive into the research topics of 'Compositional Bayesian modelling and its application to decision support in crime investigation'. Together they form a unique fingerprint.

Cite this