Identity matching using personal and social identity features

摘要: Identity verification is essential in our mission to identify potential terrorists and criminals. It not a trivial task because reportedly assume multiple identities using either fraudulent or legitimate means. A national identification card biometrics technologies have been proposed as solutions the identity problem. However, several studies show their inability tackle complex We aim develop data mining alternatives that can match referring same individual. Existing matching techniques based on primarily rely personal features. In this research, we propose new technique considers both features social define two groups of including activities relations. The built upon probabilistic relational model utilizes database structure extract Experiments activity significantly improve performance while relation effectively reduce false positive negative decisions.