A Language Modelling approach to linking criminal styles with offender characteristics

R. Bache, F. Crestani, D. Canter, D. Youngs

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

The ability to infer the characteristics of offenders from their criminal behaviour ('offender profiling') has only been partially successful since it has relied on subjective judgments based on limited data. Words and structured data used in crime descriptions recorded by the police relate to behavioural features. Thus Language Modelling was applied to an existing police archive to link behavioural features with significant characteristics of offenders. Both multinomial and multiple Bernoulli models were used. Although categories selected are gender, age group, ethnic appearance and broad occupation (employed or not), in principle this can be applied to any characteristic recorded. Results indicate that statistically significant relationships exist between all characteristics for many types of crime. Bernoulli models tend to perform better than multinomial ones. It is also possible to identify automatically specific terms which when taken together give insight into the style of offending related to a particular group.

LanguageEnglish
Pages303-315
Number of pages13
JournalData and Knowledge Engineering
Volume69
Issue number3
DOIs
Publication statusPublished - 1 Mar 2010

Fingerprint

Language modeling
Crime
Police
Profiling
Age groups

Cite this

@article{6775e6b5e11b4363b405cc909a087b73,
title = "A Language Modelling approach to linking criminal styles with offender characteristics",
abstract = "The ability to infer the characteristics of offenders from their criminal behaviour ('offender profiling') has only been partially successful since it has relied on subjective judgments based on limited data. Words and structured data used in crime descriptions recorded by the police relate to behavioural features. Thus Language Modelling was applied to an existing police archive to link behavioural features with significant characteristics of offenders. Both multinomial and multiple Bernoulli models were used. Although categories selected are gender, age group, ethnic appearance and broad occupation (employed or not), in principle this can be applied to any characteristic recorded. Results indicate that statistically significant relationships exist between all characteristics for many types of crime. Bernoulli models tend to perform better than multinomial ones. It is also possible to identify automatically specific terms which when taken together give insight into the style of offending related to a particular group.",
keywords = "Information Retrieval, Investigative psychology, Language Modelling, Offender profiling",
author = "R. Bache and F. Crestani and D. Canter and D. Youngs",
year = "2010",
month = "3",
day = "1",
doi = "10.1016/j.datak.2009.10.009",
language = "English",
volume = "69",
pages = "303--315",
journal = "Data and Knowledge Engineering",
issn = "0169-023X",
publisher = "Elsevier",
number = "3",

}

A Language Modelling approach to linking criminal styles with offender characteristics. / Bache, R.; Crestani, F.; Canter, D.; Youngs, D.

In: Data and Knowledge Engineering, Vol. 69, No. 3, 01.03.2010, p. 303-315.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Language Modelling approach to linking criminal styles with offender characteristics

AU - Bache, R.

AU - Crestani, F.

AU - Canter, D.

AU - Youngs, D.

PY - 2010/3/1

Y1 - 2010/3/1

N2 - The ability to infer the characteristics of offenders from their criminal behaviour ('offender profiling') has only been partially successful since it has relied on subjective judgments based on limited data. Words and structured data used in crime descriptions recorded by the police relate to behavioural features. Thus Language Modelling was applied to an existing police archive to link behavioural features with significant characteristics of offenders. Both multinomial and multiple Bernoulli models were used. Although categories selected are gender, age group, ethnic appearance and broad occupation (employed or not), in principle this can be applied to any characteristic recorded. Results indicate that statistically significant relationships exist between all characteristics for many types of crime. Bernoulli models tend to perform better than multinomial ones. It is also possible to identify automatically specific terms which when taken together give insight into the style of offending related to a particular group.

AB - The ability to infer the characteristics of offenders from their criminal behaviour ('offender profiling') has only been partially successful since it has relied on subjective judgments based on limited data. Words and structured data used in crime descriptions recorded by the police relate to behavioural features. Thus Language Modelling was applied to an existing police archive to link behavioural features with significant characteristics of offenders. Both multinomial and multiple Bernoulli models were used. Although categories selected are gender, age group, ethnic appearance and broad occupation (employed or not), in principle this can be applied to any characteristic recorded. Results indicate that statistically significant relationships exist between all characteristics for many types of crime. Bernoulli models tend to perform better than multinomial ones. It is also possible to identify automatically specific terms which when taken together give insight into the style of offending related to a particular group.

KW - Information Retrieval

KW - Investigative psychology

KW - Language Modelling

KW - Offender profiling

UR - http://www.scopus.com/inward/record.url?scp=74849103623&partnerID=8YFLogxK

U2 - 10.1016/j.datak.2009.10.009

DO - 10.1016/j.datak.2009.10.009

M3 - Article

VL - 69

SP - 303

EP - 315

JO - Data and Knowledge Engineering

T2 - Data and Knowledge Engineering

JF - Data and Knowledge Engineering

SN - 0169-023X

IS - 3

ER -