System and method for identity-based fraud detection through graph anomaly detection
First Claim
1. A method for identifying a fraudulent account application, comprising:
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
linking, at a fraud detection processor, the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications stored in a database, the links forming a graphical pattern; and
performing statistical analysis of the graphical pattern to determine a likelihood that the new account application is fraudulent, wherein the statistical analysis comprises analyzing the graphical pattern to determine whether the graphical pattern is anomalous when considered in relation to a normal graphical pattern.
16 Assignments
0 Petitions
Accused Products
Abstract
A method for identifying a fraudulent account application includes receiving a new account application comprising a plurality of identity-related fields and linking the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications. The links form a graphical pattern on which statistical analysis can be performed to determine the likelihood that the new account application is fraudulent. The statistical analysis can comprise comparing the graphical pattern to a known, or normal graphical pattern in order to detect differences, or anomalies occurring in the graphical pattern associated with the new account application.
-
Citations
39 Claims
-
1. A method for identifying a fraudulent account application, comprising:
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
linking, at a fraud detection processor, the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications stored in a database, the links forming a graphical pattern; and
performing statistical analysis of the graphical pattern to determine a likelihood that the new account application is fraudulent, wherein the statistical analysis comprises analyzing the graphical pattern to determine whether the graphical pattern is anomalous when considered in relation to a normal graphical pattern. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
-
9. A method for identifying a fraudulent account application, comprising:
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
linking, at a fraud detection processor, the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications stored in a database, the links forming a graphical pattern;
performing statistical analysis of the graphical pattern; and
applying a plurality of anomaly rules to the graphical pattern, wherein one of the plurality of anomaly rules is derived by analyzing graphical patterns associated with a plurality of non-fraudulent historical identity records. - View Dependent Claims (10, 11)
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
-
12. A method for identifying a fraudulent account application, comprising:
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
linking, at a fraud detection processor, the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications stored in a database, the links forming a graphical pattern;
performing statistical analysis of the graphical pattern; and
applying a plurality of anomaly rules to the graphical pattern, wherein one of the plurality of anomaly rules is derived from statistical analysis of fraud cases. - View Dependent Claims (13, 14)
- receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
-
15. In a system comprising a database of historical account applications, a method for detecting a fraudulent account application, the method comprising:
- analyzing the historical account applications stored in a database to identify a normal pattern of identity-related information and links;
developing a statistical model of the identified normal pattern;
receiving, at an identity records interface, a new account application comprising a plurality of identity-related fields;
linking, at a fraud detection processor, the identity-related fields associated with the new account application with identity-related fields associated with the historical account applications stored in the database, the links forming a graphical pattern;
performing statistical analysis of the graphical pattern based on the developed statistical model of a normal pattern; and
applying a plurality of anomaly rules to the graphical pattern. - View Dependent Claims (16, 17, 18, 19, 20, 21, 22, 23)
- analyzing the historical account applications stored in a database to identify a normal pattern of identity-related information and links;
-
24. A fraud detection system, comprising:
-
a database configured to store a plurality of historical account applications, each of the plurality of historical account applications comprising a plurality of identity-related fields; an identity records interface configured to receive a new account application comprising a plurality of identity-related fields; and a fraud detection processor interfaced with the database and the new identity records interface, the fraud detection processor configured to; link the identity-related fields associated with the new account application with identity-related fields associated with a plurality of historical account applications, the links forming a graphical pattern; perform statistical analysis of the graphical pattern; and apply a plurality of anomaly rules to the graphical pattern. - View Dependent Claims (25, 26, 27, 28, 29, 30)
-
-
31. A fraud detection system, comprising:
-
a database configured to store a plurality of historical account applications, each of the plurality of historical account applications comprising a plurality of identity-related fields; an identity records interface configured to receive a new account application comprising a plurality of identity-related fields; and a fraud detection processor interfaced with the database and the new identity records interface, the fraud detection processor configured to; analyze the historical account applications to identify a normal pattern of identity-related information and links; develop a statistical model of the identified normal pattern; link the identity-related fields associated with the new account application with identity-related fields associated with the historical account applications, the links forming a graphical pattern; perform statistical analysis of the graphical pattern based on the developed statistical model of a normal pattern; and apply a plurality of anomaly rules to the graphical pattern. - View Dependent Claims (32, 33, 34, 35, 36, 37, 38, 39)
-
Specification