The problem addressed is one of great practical significance in the investigation of stranger rape. The linkage of these crimes at an early stage is of the greatest importance in a successful prosecution and also in the prevention of further crimes that may be even more serious. The central question as it appears here relates to how these offences can be represented in a form that reflects their complexity and how, as a result, similarities can be shown and the degree of that similarity expressed within a computerised system. In this research an area that can be defined in the intersection of forensic psychology, crime analysis, decision support and artificial intelligence is considered and the possibility of a mutually enriching outcome that furthers the aim of assisting crime analysis is presented. The use of fuzzy set theory in both representing crimes and associating them in a valid manner by using fuzzy membership functions and clustering algorithms is demonstrated. This centres on the representation of each offence as a point in n-dimensional space where the dimensions are behavioural properties consisting of a number of actions or variables. Hence it is possible to express the strength of these behavioural dimensions in individual offences, to cluster them in non-exclusive sets that reflect the reality of the degrees of behaviour exhibited in the commission of crimes, and to measure the similarity between them. Consequently an extensive testing regime that varies the number of dimensions that represent crimes, the number of clusters that they are partitioned into and the ‘fuzziness’ of those clusters has been undertaken. Results indicate that, at best, the distance, or dissimilarity, between linked crimes is less than half that expected by chance. In conclusion the possibility of legitimately extending the employment of these techniques into other areas of crime analysis is discussed and the prospects of a synergy between criminal psychology and artificial intelligence examined.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:631728 |
| Date | January 2012 |
| Creators | Casey, Don |
| Publisher | London South Bank University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0016 seconds