Data breach detection

• US 10,290,001 B2
• Filed: 10/28/2014
• Issued: 05/14/2019
• Est. Priority Date: 10/28/2014
• Status: Active Grant

• Current Assignee: Brighterion Inc.
• Original Assignee: Brighterion Inc.
• Inventors: Adjaoute, Akli
• Primary Examiner(s): Niquette, Robert R
• Application Number:
  US14/525,273
• Publication Number:
  US 20150073981A1
• Time in Patent Office: 1,659 Days
• CPC Class Codes: G06Q 20/384 using social networks
• CPC Class Codes: G06Q 20/4016 involving fraud or risk level assessment in transaction processing
• CPC Class Codes: G06Q 30/0225 Avoiding frauds

Abstract

A merchant data breach process comprises processing daily payment transaction data with a risk and compliance platform to obtain a fraud score for each constituent transaction. Constituent transactions with high risk fraud scores are sorted into a table according to the transaction date, cardholder, and merchant. The table data is scored according to suspected card visits, highly probable visits, and all card visits. The scores are normalized according to merchant size grouping through the use of multipliers. The normalized scores are summed together day-by-day into a final score. A timely warning of an underlying and expanding security rupture caused by a merchant data breach is issued for damage control and law enforcement.

Claims

1. A data processing system that is artificial intelligence based and computer implemented for the detection of merchant data breaches, comprising:

means that receives an inflow of payment transaction data over a network comprising merchant identifiers and visiting payment card identifiers into a risk and compliance platform part of a data processing system;

means that computes a fraud score for each constituent transaction in a daily run of said payment transaction data with said risk and compliance platform;

means that marks any said constituent transactions with fraud scores computed to exceed a first threshold, and that sorts these into a table data structure according to a transaction date, a cardholder, and a merchant;

means that tallies for each corresponding merchant in said constituent transactions a number of payment card visits apparent in the daily payment transaction data that received a fraud score exceeding said first threshold and thereby classified as suspect or suspicious, and that tallies a total number for all payment card visits in said constituent transactions;

means that scores any data tallied in the table data structure according to a number of payment card visits that are suspected or at least highly probable of fraud, and that computes a total of all payment card visits;

means that standardizes and normalizes the fraud scores of any payment card visits according to a score-balancing multiplication factor applied to each of many merchant size groupings into a consistent fraud score;

means that sums said consistent fraud scores together into a daily final fraud score;

means that issues particular warnings of a merchant data breach if any of said daily final fraud scores exceeds a second threshold value; and

means that issues a general warning of an underlying and expanding security rupture computed to stem in common from a merchant data breach.

2. The data processing system for the detection of merchant data breaches of claim 1, further comprising:

means that transforms fraud scores by computing the square root of a number of payment card visits suspected to be fraudulent due to a particular fraud score; and

means that multiplies a resulting square root of said number of payment card visits suspected to be fraudulent by said score-balancing multiplication factor, and thereby produces said consistent fraud scores that are normalized and standardized.

3. The data processing system for the detection of merchant data breaches of claim 1, further comprising:

means that transforms fraud scores by computing the square root of a number of payment card visits suspected to be fraudulent due to a particular fraud score divided by all payment card visits; and

means that multiplies a result by said score-balancing multiplication factor, and thereby produces said consistent fraud scores that are normalized and standardized.

4. The data processing system for the detection of merchant data breaches of claim 1, further comprising:

means that transforms fraud scores by computing the square root of the number of payment card visits probable to be fraudulent due to a particular fraud score divided by all card visits; and

means that multiplies a result by said score-balancing multiplication factor, and thereby produces said consistent fraud scores that are normalized and standardized.

5. The data processing system for the detection of merchant data breaches of claim 1, further comprising:

means that groups said merchant identifiers into size categories according to a number of payment card visits attributable to each.

6. The data processing system for the detection of merchant data breaches of claim 1, further comprising:

means that generates within the data processing system an initial population of smart agent profile data structures data mined from historical transaction data;

means that compares each newly arriving transaction record in said inflow of payment transaction data attribute-by-attribute with a time interval series of attributes in a corresponding smart agent profile data structure, and that incrementally increases or decreases a computed fraud risk score; and

means that outputs said computed fraud risk score as a determination an underlying transaction is genuine, suspicious, or fraudulent.

7. The data processing system for the detection of merchant data breaches of claim 6, further comprising:

means that divides each said time interval series into a real-time part and a long-term part; and

means that pre-computes a velocity count and statistics for each said real-time part and long-term part of said individual and expanded attributes;

means that compares each said newly arriving transaction record item-by-item to relevant items in each said real-time part and long-term part, and thereby determines if each item represents known behavior or unknown behavior.

8. The data processing system for the detection of merchant data breaches of claim 6, further comprising:

means that inspects each said newly arriving transaction record for a corresponding smart agent profile data structure; and

means that generates a smart agent profile data structure for entities detected as not already having one associated.

9. The data processing system for the detection of merchant data breaches of claim 6, further comprising:

means that generates three populations of smart agent profile data structures data mined from historical transaction data according to a corresponding number of cardholder, merchant, and identified device entities involved in each transaction represented in said payment transaction data; and

means that compares each newly arriving transaction record in said inflow of payment transaction data attribute-by-attribute with a time interval series of attributes in a smart agent profile data structure corresponding to a particular cardholder, a particular merchant, and a particular identified device, and that incrementally increases or decreases a computed fraud risk score.