Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes

Richard Bache; Fabio Crestani; David Canter; Donna Youngs

doi:10.1109/IAS.2007.58

Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes

Richard Bache, Fabio Crestani, David Canter, Donna Youngs

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Citations (Scopus)

Abstract

Language Models are successfully applied to the problem of analysing crime descriptions from a police database with the purpose of prioritising suspects for an unsolved crime, given details of solved crimes. The frequency of terms in each description relates to the behaviour of the offender and this can be used to link crimes to a common offender. Language Modelling uses Bayes' theorem and thus require a prior probability. Such a prior can be based on each offender's past propensity to offend, derived from historic data. Language Modelling yields a probability of a document being relevant, which in this case is interpreted as the probability of a suspect being the culprit. Although the absolute value of the probability does not carry any direct applied implications, the study does show that the general likelihood of identification of the actual suspect does correspond to the relative values. Thus these probabilities can be used for more than just ranking suspects.

Original language	English
Title of host publication	Third International Symposium on Information Assurance and Security
Editors	Ning Zhang, Ajith Abraham, Qi Shi, Johnson Thomas
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	399-404
Number of pages	6
ISBN (Electronic)	9780769528762
ISBN (Print)	0769528767
DOIs	https://doi.org/10.1109/IAS.2007.58
Publication status	Published - 10 Sep 2007
Externally published	Yes
Event	3rd International Symposium on Information Assurance and Security - Manchester, United Kingdom Duration: 29 Aug 2007 → 31 Aug 2007 Conference number: 3 https://ieeexplore.ieee.org/document/4299734

Conference

Conference	3rd International Symposium on Information Assurance and Security
Abbreviated title	IAS 2007
Country/Territory	United Kingdom
City	Manchester
Period	29/08/07 → 31/08/07
Internet address	https://ieeexplore.ieee.org/document/4299734

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/IAS.2007.58Licence: Unspecified

Cite this

Bache, R., Crestani, F., Canter, D., & Youngs, D. (2007). Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes. In N. Zhang, A. Abraham, Q. Shi, & J. Thomas (Eds.), Third International Symposium on Information Assurance and Security (pp. 399-404). [4299806] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IAS.2007.58

@inproceedings{10d97f6b1b034bf7b34083d048ce8fa8,

title = "Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes",

abstract = "Language Models are successfully applied to the problem of analysing crime descriptions from a police database with the purpose of prioritising suspects for an unsolved crime, given details of solved crimes. The frequency of terms in each description relates to the behaviour of the offender and this can be used to link crimes to a common offender. Language Modelling uses Bayes' theorem and thus require a prior probability. Such a prior can be based on each offender's past propensity to offend, derived from historic data. Language Modelling yields a probability of a document being relevant, which in this case is interpreted as the probability of a suspect being the culprit. Although the absolute value of the probability does not carry any direct applied implications, the study does show that the general likelihood of identification of the actual suspect does correspond to the relative values. Thus these probabilities can be used for more than just ranking suspects.",

keywords = "Information retrieval, Databases, Frequency, Psychology, Data security, Information security, Computer crime, Information science, Information analysis, Data analysis",

author = "Richard Bache and Fabio Crestani and David Canter and Donna Youngs",

year = "2007",

month = sep,

day = "10",

doi = "10.1109/IAS.2007.58",

language = "English",

isbn = "0769528767",

pages = "399--404",

editor = "Zhang, {Ning } and Abraham, {Ajith } and Qi Shi and Thomas, {Johnson }",

booktitle = "Third International Symposium on Information Assurance and Security",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "3rd International Symposium on Information Assurance and Security, IAS 2007 ; Conference date: 29-08-2007 Through 31-08-2007",

url = "https://ieeexplore.ieee.org/document/4299734",

}

Bache, R, Crestani, F, Canter, D & Youngs, D 2007, Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes. in N Zhang, A Abraham, Q Shi & J Thomas (eds), Third International Symposium on Information Assurance and Security., 4299806, Institute of Electrical and Electronics Engineers Inc., pp. 399-404, 3rd International Symposium on Information Assurance and Security, Manchester, United Kingdom, 29/08/07. https://doi.org/10.1109/IAS.2007.58

Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes. / Bache, Richard; Crestani, Fabio; Canter, David et al.

Third International Symposium on Information Assurance and Security. ed. / Ning Zhang; Ajith Abraham; Qi Shi; Johnson Thomas. Institute of Electrical and Electronics Engineers Inc., 2007. p. 399-404 4299806.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes

AU - Bache, Richard

AU - Crestani, Fabio

AU - Canter, David

AU - Youngs, Donna

N1 - Conference code: 3

PY - 2007/9/10

Y1 - 2007/9/10

N2 - Language Models are successfully applied to the problem of analysing crime descriptions from a police database with the purpose of prioritising suspects for an unsolved crime, given details of solved crimes. The frequency of terms in each description relates to the behaviour of the offender and this can be used to link crimes to a common offender. Language Modelling uses Bayes' theorem and thus require a prior probability. Such a prior can be based on each offender's past propensity to offend, derived from historic data. Language Modelling yields a probability of a document being relevant, which in this case is interpreted as the probability of a suspect being the culprit. Although the absolute value of the probability does not carry any direct applied implications, the study does show that the general likelihood of identification of the actual suspect does correspond to the relative values. Thus these probabilities can be used for more than just ranking suspects.

AB - Language Models are successfully applied to the problem of analysing crime descriptions from a police database with the purpose of prioritising suspects for an unsolved crime, given details of solved crimes. The frequency of terms in each description relates to the behaviour of the offender and this can be used to link crimes to a common offender. Language Modelling uses Bayes' theorem and thus require a prior probability. Such a prior can be based on each offender's past propensity to offend, derived from historic data. Language Modelling yields a probability of a document being relevant, which in this case is interpreted as the probability of a suspect being the culprit. Although the absolute value of the probability does not carry any direct applied implications, the study does show that the general likelihood of identification of the actual suspect does correspond to the relative values. Thus these probabilities can be used for more than just ranking suspects.

KW - Information retrieval

KW - Databases

KW - Frequency

KW - Psychology

KW - Data security

KW - Information security

KW - Computer crime

KW - Information science

KW - Information analysis

KW - Data analysis

UR - http://www.scopus.com/inward/record.url?scp=46749151832&partnerID=8YFLogxK

U2 - 10.1109/IAS.2007.58

DO - 10.1109/IAS.2007.58

M3 - Conference contribution

AN - SCOPUS:46749151832

SN - 0769528767

SP - 399

EP - 404

BT - Third International Symposium on Information Assurance and Security

A2 - Zhang, Ning

A2 - Abraham, Ajith

A2 - Shi, Qi

A2 - Thomas, Johnson

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd International Symposium on Information Assurance and Security

Y2 - 29 August 2007 through 31 August 2007

ER -

Application of Language Models to Suspect Prioritisation and Suspect Likelihood in Serial Crimes

Abstract

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this