Method and system for data security utilizing user behavior and device identification
AssigneeVisa International Service Association
InventorsB. Scott Boding, Andrew Naumann zu Koenigsbrueck
Life of this patent
Estimating the term…
Abstract
Embodiments of the invention broadly described, introduce systems and methods for detecting the likelihood that a transaction is fraudulent using user access pattern data and device fingerprint data. One embodiment of the invention discloses a method for determining the likelihood that a payment transaction is fraudulent. The method comprises receiving user access pattern data generated by a user client computer, receiving a device fingerprint associated with the user client computer conducting a payment transaction, determining, using the user access pattern data and the device fingerprint, a likelihood that the payment transaction is fraudulent.
Claims20
A computer-implemented method comprising:
- receiving, by a processor over a network, user access pattern data generated by a user client computer, wherein the user client computer is used to request an authentication process, wherein the user access pattern data corresponds to user interactions of a user with one or more input devices of the user client computer, the user interactions detected by the one or more input devices;
- receiving, by the processor over the network, device fingerprint data associated with the user client computer;
- determining, by the processor, a type of device for the user client computer by using the device fingerprint data to access a list of device types;
- calculating, by the processor, a value of user interactions with the user client computer based on the user access pattern data;
- determining, by the processor, a likelihood that the request is fraudulent by comparing the value to a threshold;
- and
- declining, by the processor, authentication of the user if the likelihood indicates that the user access pattern data is inconsistent with the type of device determined based on the device fingerprint data.
A transaction analysis system, comprising:
- a processor;
- and
- a non-transitory computer-readable storage medium, comprising code executable by the processor for implementing a method comprising:
- receiving, over a network, user access pattern data generated by a user client computer, wherein the user client computer is used to request an authentication process, wherein the user access pattern data corresponds to user interactions of a user with one or more input devices of the user client computer, the user interactions detected by the one or more input devices;
- receiving, over the network, device fingerprint data associated with the user client computer;
- determining a type of device for the user client computer by using the device fingerprint data to access a list of device types;
- calculating a value of user interactions with the user client computer based on the user access pattern data;
- determining a likelihood that the request is fraudulent by comparing the value to a threshold;
- and
- declining authentication of the user if the likelihood indicates that the user access pattern data is inconsistent with the type of device determined based on the device fingerprint data.
A non-transitory computer-readable storage medium, comprising code executable by a processor for implementing a method comprising:
- receiving, over a network, user access pattern data generated by a user client computer, wherein the user client computer is used to request an authentication process, wherein the user access pattern data corresponds to user interactions of a user with one or more input devices of the user client computer, the user interactions detected by the one or more input devices;
- receiving, over the network, device fingerprint data associated with the user client computer;
- determining a type of device for the user client computer by using the device fingerprint data to access a list of device types;
- calculating a value of user interactions with the user client computer based on the user access pattern data;
- determining a likelihood that the request is fraudulent by comparing the value to a threshold;
- and
- declining authentication of the user if the likelihood indicates that the user access pattern data is inconsistent with the type of device determined based on the device fingerprint data.
Dependencies are inferred from the claim text.