Estimating real-valued characteristics of criminals from their recorded crimes

ABSTRACT

Offender profiling concerns making inferences about a criminal from the crime(s) he has committed. Where descriptionsof the crimes are recorded electronically, text mining techniques provide a means by which recorded characteristics of the offenders can be linked with features of his crimes as revealed in the text. Past studies have used Language Modelling to identify characteristics that can be described by a categorical variable e.g. gender. Here we adapt the Language Modelling approach to allow estimation of numerical quantities such as age and distance travelled.

REFERENCES

• 1: Richard Bache , Fabio Crestani , David Canter , Donna Youngs, A Language Modelling Approach to Linking Criminal Styles with Offender Characteristics, Proceedings of the 13th international conference on Natural Language and Information Systems: Applications of Natural Language to Information Systems, June 24-27, 2008, London, UK [doi>10.1007/978-3-540-69858-6_26]
• 2: Richard Bache , Fabio Crestani, Towards an Automated Approach to Offender Profiling, Proceedings of the 2008 International Conference on Computational Sciences and Its Applications, p.537-545, June 30-July 03, 2008 [doi>10.1109/ICCSA.2008.35]
• 3: D. Canter & K. Fritzon, Differentiating arsonists: A Model of Firesetting Actions and Characteristics, Legal and Criminal Psychology, vol. 3, pp 73--96, 1998.
• 4: C. Fellbaum (Ed.): WordNet - An Electronic Lexical Database, MIT Press, 1998.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

General Terms:
Algorithms, Experimentation, Measurement, Theory

Keywords:
bayesian analysis, crime data, data mining, language modelling, offender profiling