System for identifying points of compromise
AssigneeU.S. Bank National Association
InventorsChristopher Kallas, Xiaoqiao Wei
Life of this patent
Estimating the term…
Abstract
The disclosure describes an apparatus having programmed instructions that when executed cause the apparatus to receive, via a communication network, information regarding suspicious fraud activity at a first location involving a plurality of transaction cards; monitor changes over a first time interval to received information regarding suspicious fraud activity at the first location; and identify a point-of-compromise (POC) location based on monitored changes surpassing a threshold indicating suspicious fraud activity at the first location over the first time interval.
Claims20
A method comprising:
- receiving, by a computing system comprising a hierarchical memory system comprising a first tier and a second tier, the first tier having a faster storage technology relative to a second storage technology of the second tier, transaction data from a plurality of point-of-sale terminals;
- storing, from the transaction data, recent transaction data and post-processed indicators in the first tier;
- storing, from the transaction data, historical transaction data in the second tier;
- maintaining, in the first tier by the computing system, a three-dimensional matrix comprising a first dimension representing transaction cards, a second dimension representing merchant locations, and a third dimension representing time intervals;
- determining, by the computing system, a compromise event comprising a suspicious fraud activity, based on a pattern in the three-dimensional matrix by monitoring a transaction card that performed a non-fraudulent transaction at a merchant location during a time interval and subsequently performed a fraudulent transaction at another location;
- identifying, by the computing system, a candidate point-of-compromise (POC) location based on the suspicious fraud activity;
- generating, by the computing system, a mass count plot showing a number of transaction cards having suspicious fraud activity at the candidate POC location for each of a plurality of time intervals;
- identifying, by the computing system, a compromise time window within the mass count plot;
- and
- generating, by the computing system, a reissue score for a transaction card used at the candidate POC location during the compromise time window.
A method comprising:
- receiving, by a computing system, suspicious fraud activity data for a plurality of transaction cards;
- storing the suspicious fraud activity data in a static random access memory (SRAM) array, wherein rows of the memory array correspond to transaction cards, columns of the memory array correspond to a plurality of merchant locations, and a cell of the memory array stores a counter value indicating a level of suspicious fraud activity for a transaction card at a merchant location in the plurality of merchant locations;
- identifying, by the computing system, a candidate point-of-compromise (POC) location based on a distribution pattern of a set of counter values in the memory array;
- performing, by the computing system, a verification check on the candidate POC location comprising:
- analyzing a geographic clustering of a set of fraudulent transactions that occurred subsequent to a transaction at the candidate POC location,
- comparing with a known compromise event time pattern a time pattern of a fraudulent transaction in the set of fraudulent transactions;
- and
- correlating a characteristic of the fraudulent transaction with a merchant-specific transaction characteristic at the candidate POC location;
- and
- assigning, by the computing system, a confidence level to the candidate POC location based on a result of the verification check.
Dependencies are inferred from the claim text.