Identity matching based on probabilistic relational models

Jiexun Li, Gang Wang, Hsinchun Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Identity management is critical to various organizational practices ranging from citizen services to crime investigation. The task of searching for a specific identity is difficult because multiple identity representations may exist due to issues related to unintentional errors and intentional deception. In this study we propose a probabilistic relational model (PRM) based approach to match identities in databases. By exploring a database relational structure, we derive three categories of features, namely personal identity features, social activity features, and social relationship features. Based on these derived features, a probabilistic prediction model can be constructed to make a matching decision on a pair of identities. An experimental study using a real criminal dataset demonstrates the effectiveness of the proposed PRM-based approach. By incorporating social activity features, the average precision of identity matching increased from 53.73 % to 54.64%; furthermore, the incorporation of social relation features increased the average precision to 68.27%.

Original languageEnglish (US)
Title of host publicationAssociation for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006
Pages1457-1464
Number of pages8
Volume3
StatePublished - 2006
Event12th Americas Conference on Information Systems, AMCIS 2006 - Acapulco, Mexico
Duration: Aug 4 2006Aug 6 2006

Other

Other12th Americas Conference on Information Systems, AMCIS 2006
CountryMexico
CityAcapulco
Period8/4/068/6/06

Fingerprint

Crime
Social Relations
offense
citizen
management

Keywords

  • Feature construction
  • Identity matching
  • Probabilistic relational models

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Networks and Communications
  • Library and Information Sciences
  • Information Systems

Cite this

Li, J., Wang, G., & Chen, H. (2006). Identity matching based on probabilistic relational models. In Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006 (Vol. 3, pp. 1457-1464)

Identity matching based on probabilistic relational models. / Li, Jiexun; Wang, Gang; Chen, Hsinchun.

Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006. Vol. 3 2006. p. 1457-1464.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, J, Wang, G & Chen, H 2006, Identity matching based on probabilistic relational models. in Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006. vol. 3, pp. 1457-1464, 12th Americas Conference on Information Systems, AMCIS 2006, Acapulco, Mexico, 8/4/06.
Li J, Wang G, Chen H. Identity matching based on probabilistic relational models. In Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006. Vol. 3. 2006. p. 1457-1464
Li, Jiexun ; Wang, Gang ; Chen, Hsinchun. / Identity matching based on probabilistic relational models. Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006. Vol. 3 2006. pp. 1457-1464
@inproceedings{d1ce19adb00d467f917db2ab29d0a019,
title = "Identity matching based on probabilistic relational models",
abstract = "Identity management is critical to various organizational practices ranging from citizen services to crime investigation. The task of searching for a specific identity is difficult because multiple identity representations may exist due to issues related to unintentional errors and intentional deception. In this study we propose a probabilistic relational model (PRM) based approach to match identities in databases. By exploring a database relational structure, we derive three categories of features, namely personal identity features, social activity features, and social relationship features. Based on these derived features, a probabilistic prediction model can be constructed to make a matching decision on a pair of identities. An experimental study using a real criminal dataset demonstrates the effectiveness of the proposed PRM-based approach. By incorporating social activity features, the average precision of identity matching increased from 53.73 {\%} to 54.64{\%}; furthermore, the incorporation of social relation features increased the average precision to 68.27{\%}.",
keywords = "Feature construction, Identity matching, Probabilistic relational models",
author = "Jiexun Li and Gang Wang and Hsinchun Chen",
year = "2006",
language = "English (US)",
isbn = "9781604236262",
volume = "3",
pages = "1457--1464",
booktitle = "Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006",

}

TY - GEN

T1 - Identity matching based on probabilistic relational models

AU - Li, Jiexun

AU - Wang, Gang

AU - Chen, Hsinchun

PY - 2006

Y1 - 2006

N2 - Identity management is critical to various organizational practices ranging from citizen services to crime investigation. The task of searching for a specific identity is difficult because multiple identity representations may exist due to issues related to unintentional errors and intentional deception. In this study we propose a probabilistic relational model (PRM) based approach to match identities in databases. By exploring a database relational structure, we derive three categories of features, namely personal identity features, social activity features, and social relationship features. Based on these derived features, a probabilistic prediction model can be constructed to make a matching decision on a pair of identities. An experimental study using a real criminal dataset demonstrates the effectiveness of the proposed PRM-based approach. By incorporating social activity features, the average precision of identity matching increased from 53.73 % to 54.64%; furthermore, the incorporation of social relation features increased the average precision to 68.27%.

AB - Identity management is critical to various organizational practices ranging from citizen services to crime investigation. The task of searching for a specific identity is difficult because multiple identity representations may exist due to issues related to unintentional errors and intentional deception. In this study we propose a probabilistic relational model (PRM) based approach to match identities in databases. By exploring a database relational structure, we derive three categories of features, namely personal identity features, social activity features, and social relationship features. Based on these derived features, a probabilistic prediction model can be constructed to make a matching decision on a pair of identities. An experimental study using a real criminal dataset demonstrates the effectiveness of the proposed PRM-based approach. By incorporating social activity features, the average precision of identity matching increased from 53.73 % to 54.64%; furthermore, the incorporation of social relation features increased the average precision to 68.27%.

KW - Feature construction

KW - Identity matching

KW - Probabilistic relational models

UR - http://www.scopus.com/inward/record.url?scp=84870208179&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870208179&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781604236262

VL - 3

SP - 1457

EP - 1464

BT - Association for Information Systems - 12th Americas Conference On Information Systems, AMCIS 2006

ER -