Automatically detecting deceptive criminal identities

Gang Wang, Hsinchun Chen, Homa Atabakhsh

Research output: Contribution to journalArticle

70 Scopus citations


The uncovering patterns of criminal identity deception based on actual criminal records and algorithmic approach to reveal deceptive identities are discussed. The testing results shows that no false positive errors occurs which shows the effectiveness of the algorithm. The errors occurs in the false negative category in which unrelated suspects are recognized as being related. The threshold value is set to capture maximum possible true similar records. Adaptive threshold is required for making an automated process in the future research.

Original languageEnglish (US)
Pages (from-to)70-76
Number of pages7
JournalCommunications of the ACM
Issue number3
Publication statusPublished - Mar 2004


ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this