Methods and systems for automatically configuring user authentication rules

• US 10,650,128 B2
• Filed: 10/18/2017
• Issued: 05/12/2020
• Est. Priority Date: 10/18/2017
• Status: Active Grant

• Current Assignee: Mastercard International Incorporated (MasterCard Incorporated)
• Original Assignee: Mastercard International Incorporated (MasterCard Incorporated)
• Inventors: Nowak, Dawid, Waldron, Ashley, Nicoletti, Laurie
• Primary Examiner(s): Hirl, Joseph P
• Assistant Examiner(s): Gyorfi, Thomas A
• Application Number:
  US15/787,097
• Publication Number:
  US 20190114404A1
• Time in Patent Office: 937 Days
• CPC Class Codes: G06F 21/31 User authentication
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/042 Knowledge-based neural networks; Logical representations of neural networks
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: H04L 63/08 for authentication of entities cryptographic mechanisms or cryptographic arrangements for entity authentication H04L9/32
• CPC Class Codes: H04L 63/102 Entity profiles
• CPC Class Codes: H04L 63/205 involving negotiation or determination of the one or more network security mechanisms to be used, e.g. by negotiation between the client and the server or between peers or by selection according to the capabilities of the entities involved negotiation of communication capabilities H04L69/24
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04W 12/06 Authentication
• CPC Class Codes: H04W 12/37 Managing security policies for mobile devices or for controlling mobile applications

Abstract

Systems and processes for automatically configuring user authentication rules for each of a plurality of users for use in transactions. A neural network engine receives first party preferences data from a first party computer that includes user authentication requirement criteria associated with a plurality of transaction types, and receives at least two of user behavior data, user historical data, authenticator data associated with a mobile device of the user, and mobile device metadata. The neural network engine then generates an output value based on this data, transmits the output value to a score comparator for comparison to a required score specified by the first party, and receives feedback data from the score comparator when the output value is not within a tolerance of the required score. When the output value is within the tolerance, then the neural network engine generates user authentication rules recommendations and transmits them to the first party computer.

Claims

1. A method for automatically determining user authentication criteria, comprising:

receiving, by a neural network engine from a first party computer, first party preferences data comprising user authentication requirement criteria associated with a plurality of different transaction types, the user authentication criteria comprising a fraudulent transaction tolerance associated with each transaction type of the plurality of different transaction types;

receiving, by the neural network engine, authenticator data associated with a mobile device of a user, mobile device metadata, and at least one of user behavior data and user historical data;

generating, by the neural network engine, an output value based on the first party preferences data, the authenticator data associated with the mobile device of the user, the mobile device metadata, and at least one of the user behavior data and the user historical data;

transmitting, by the neural network engine to a score comparator, the output value for comparison to a required score specified by the first party;

receiving, by the neural network engine from the score comparator, feedback data when the output value is not within a tolerance of the required score;

generating, by the neural network engine, an updated output value based on the feedback data;

transmitting, by the neural network engine to the score comparator, the updated output value for comparison to the required score;

receiving, by the neural network engine from the score comparator, a match indication;

generating, by the neural network engine, user authentication rules recommendations associated with the user for each of the plurality of different transaction types; and

transmitting, by the neural network engine, the user authentication rules recommendations to the first party computer.

2. The method of claim 1, wherein the updated output value is generated by the neural network engine at regular intervals.

3. The method of claim 1, wherein the user authentication rules recommendations comprise a list of authenticators used to create a model for the user associated with the output value closest to the required score.

4. The method of claim 1, wherein at least one of the first party preferences data, the user behavior data, the user historical data, the authenticator data, and the mobile device metadata is unstructured data.

5. A user authentication rules recommendation system comprising:

a user mobile device comprising at least one authenticator;

a first party computer; and

a rules recommender computer system operably connected to the first party computer and in communication with the user mobile device, the rules recommender computer system comprising;

a neural network engine operably connected to a plurality of classifiers;

a score comparator operably connected to the neural network engine; and

a first party rules engine operably connected to the score comparator;

and wherein the neural network engine of the rules recommendation computer system is configured to;

receive first party preferences from the first party computer, the first party preferences data comprising user authentication requirement criteria associated with a plurality of different transaction types, the user authentication criteria comprising a fraudulent transaction tolerance associated with each transaction type of the plurality of different transaction types;

receive authenticator data associated with a mobile device of the user, mobile device metadata, and at least one of user behavior data and user historical data;

generate an output value based on the first party preferences data, the authenticator data associated with the mobile device of the user, the mobile device metadata, and at least one of the user behavior data and the user historical data;

transmit the output value to the score comparator for comparison to a required score specified by the first party;

receive feedback data from the score comparator when the output value is not within a tolerance of the required score;

generate an updated output value based on the feedback data;

transmit the updated output value to the score comparator for comparison to the required score;

receive a match indication from the score comparator;

generate user authentication rules recommendations associated with the user for each of the plurality of different transaction types; and

transmit the user authentication rules recommendations to the first party computer.

6. The system of claim 5, wherein the at least one authenticator comprises at least one of a digital camera, a fingerprint reader, a touch screen, and a microphone for recording voice data.

7. The system of claim 5, wherein the neural network engine is further configured generate the updated output at regular intervals.

8. The system of claim 5, wherein neural network engine is configured to transmit user authentication rules recommendations comprising a list of authenticators used to create a model for the user associated with the output value closest to the required score.