Patent Issued for Reducing false positives using customer data and machine learning (USPTO 11049109)

2021 JUL 15 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- State Farm Mutual Automobile Insurance Company (Bloomington, Illinois, United States) has been issued patent number 11049109, according to news reporting originating out of Alexandria, Virginia, by NewsRx editors.

The patent’s inventors are Batra, Reena (Alpharetta, GA, US), Craig, Bradley A. (Normal, IL, US), Dua, Puneit (Bloomington, IL, US), Flowers, Elizabeth A. (Bloomington, IL, US), Kramme, Timothy (Parker, TX, US), Phillips, Shanna L. (Bloomington, IL, US), Ruestman, Russell (Minonk, IL, US), Valero, Miriam (Bloomington, IL, US).

This patent was filed on June 11, 2020 and was published online on June 29, 2021.

From the background information supplied by the inventors, news correspondents obtained the following quote: “Financial fraud, in its many forms, is a problem of enormous magnitude and scope, causing billions of dollars in economic losses and impacting many millions of people. Types of financial fraud include use of a lost or stolen card, account takeover, skimming, chargeback (“friendly”) fraud, counterfeiting, forgeries and application (e.g., loan application) fraud, to name just a few. The problem only continues to grow as various technological advances, intended to improve convenience and efficiency in the marketplace, provide new opportunities for bad actors. For example, an ever-increasing amount of fraud may be linked to online transactions made via the Internet.

“Various software applications have been developed to detect potentially fraudulent transactions. For example, dollar amounts and geographic locations have generally been used to flag particular credit or debit card transactions, with cardholders then being contacted by employees of the card issuer to determine whether the transactions were indeed fraudulent. To ensure that most instances of fraud are captured, however, such techniques generally have a low threshold for triggering a fraud alert. As a result, numerous fraud alerts are false positives. The prevalence of false positives leads to a large cost in terms of the drain on human resources (e.g., calling customers to discuss each suspect transaction, and/or other manual investigation techniques), and considerable distraction or annoyance for cardholders. To provide a solution to these shortcomings in the field of automated fraud detection, innovative processing techniques capable of reducing false positives are needed.”

Supplementing the background information on this patent, NewsRx reporters also obtained the inventors’ summary information for this patent: “The present embodiments may, inter alia, reduce false positive fraud alerts using customer data. For example, fraud may be detected, verified and/or classified using customer locations, customer online activity, customer preferences, and/or other information. Moreover, in some embodiments, the rules used to detect, verify and/or classify fraud may be generated by a machine learning program. For example, supervised learning techniques may be used, with the machine learning program analyzing various types of data (e.g., including any of the data types listed above) associated with financial transactions, financial accounts and/or account holders in confirmed cases of fraud to determine which information is most probative of fraudulent activity or the lack thereof.

“In one embodiment, a computer-implemented method of detecting whether electronic fraud alerts are false positives prior to transmission to customer mobile devices based upon customer data includes: (1) receiving, by one or both of (i) one or more processors and (ii) one or more transceivers, data detailing a financial transaction associated with a customer, the data detailing the financial transaction being transmitted from a merchant computer terminal over one or more radio links; (2) inputting, by the one or more processors, the data detailing the financial transaction into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data detailing the financial transaction; (3) when an electronic fraud alert is generated for the financial transaction, inputting, by the one or more processors, the data detailing the financial transaction into a machine learning program that is trained to identify one or more facts indicated by the data detailing the financial transaction that caused the electronic fraud alert to be generated; (4) determining, by the one or more processors, whether the identified one or more facts that caused the electronic fraud alert to be generated can be verified by customer data; (5) in response to determining that the identified one or more facts that caused the electronic fraud alert to be generated can be verified by customer data, retrieving or receiving, by the one or more processors, first customer data; (6) verifying, by the one or more processors, that the electronic fraud alert is not a false positive based upon analysis of the first customer data; and/or (7) transmitting, by one or both of (i) the one or more processors and (ii) the one or more transceivers, the verified electronic fraud alert to a mobile device of the customer via a wireless communication channel to alert the customer to fraudulent activity. The method may include additional, less, or alternate actions, including those discussed elsewhere herein.

“In another embodiment, a computer-implemented method of detecting whether electronic fraud alerts are false positives prior to transmission to customer mobile devices based upon customer data includes: (1) receiving, by one or both of (i) one or more processors and (ii) one or more transceivers, data detailing a financial transaction associated with a customer, the data detailing the financial transaction being transmitted from a merchant computer terminal over one or more radio links; (2) inputting, by the one or more processors, the data detailing the financial transaction into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data detailing the financial transaction; (3) when an electronic fraud alert is generated for the financial transaction, inputting, by the one or more processors, the data detailing the financial transaction into a machine learning program that is trained to one or both of (i) determine a reason why the electronic fraud alert was generated, and (ii) identify one or more facts indicated by the data detailing the financial transaction that caused the electronic fraud alert to be generated; (4) determining, by the one or more processors, whether one or both of (i) the reason why the electronic fraud alert was generated, and (ii) the identified one or more facts that caused the electronic fraud alert to be generated, can be verified by customer data; (5) in response to determining that one or both of (i) the reason why the electronic fraud alert was generated, and (ii) the identified one or more facts that caused the electronic fraud alert to be generated, can be verified by customer data, retrieving or receiving, by the one or more processors, first customer data; (6) verifying, by the one or more processors, that the electronic fraud alert is not a false positive based upon analysis of the first customer data; and/or (7) transmitting, by one or both of (i) the one or more processors and (ii) the one or more transceivers, the verified electronic fraud alert to a mobile device of the customer via a wireless communication channel to alert the customer to fraudulent activity. The method may include additional, less, or alternate actions, including those discussed elsewhere herein.

“In another embodiment, a computer system configured to detect whether electronic fraud alerts are false positives prior to transmission to customer mobile devices based upon customer data includes one or both of (i) one or more processors and (ii) one or more transceivers, and is configured to: (1) receive data detailing a financial transaction associated with a customer, the data detailing the financial transaction being transmitted from a merchant computer terminal over one or more radio links; (2) input the data detailing the financial transaction into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data detailing the financial transaction; (3) when an electronic fraud alert is generated for the financial transaction, input the data detailing the financial transaction into a machine learning program that is trained to identify one or more facts indicated by the data detailing the financial transaction that caused the electronic fraud alert to be generated; (4) determine whether the identified one or more facts that caused the electronic fraud alert to be generated can be verified by customer data; (5) in response to determining that the identified one or more facts that caused the electronic fraud alert to be generated can be verified by customer data, retrieve or receive first customer data; (6) verify that the electronic fraud alert is not a false positive based upon analysis of the first customer data; and/or (7) transmit the verified electronic fraud alert to a mobile device of the customer via wireless communication channel to alert the customer of fraudulent activity and to facilitate not transmitting false positives to customer mobile devices. The computer system may include additional, less, or alternate functionality, including that discussed elsewhere herein.”

The claims supplied by the inventors are:

“1. A computer-implemented method of detecting whether electronic fraud alerts are false positives, the method comprising: training a first machine learning program using fraud classifications made in connection with at least one of a type of transaction data or a value of transaction data associated with a plurality of financial accounts, such that the first machine learning program learns a characteristic of the transaction data that is indicative of different fraud classifications; receiving, by one or more processors, transaction data detailing a financial transaction associated with a customer; inputting, by the one or more processors, the data detailing the financial transaction into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data detailing the financial transaction; when an electronic fraud alert is generated for the financial transaction, inputting, by the one or more processors, the data detailing the financial transaction into the first machine learning program, and determining, using the trained first machine learning program and based at least in part on a type of the transaction data or a value of the transaction data, a reason why the electronic fraud alert was generated, determining, by the one or more processors, that the reason why the electronic fraud alert was generated can be verified by customer data; verifying, by the one or more processors, that the electronic fraud alert is not a false positive based upon analysis of first customer data; and transmitting, by one or more transceivers, the electronic fraud alert to a mobile device of the customer.

“2. The computer-implemented method of claim 1, wherein the first customer data is associated with one or both of (i) a date of the financial transaction, and (ii) a time of the financial transaction.

“3. The computer-implemented method of claim 1, wherein the first customer data is associated with, or generated and transmitted by, a customer vehicle or a customer mobile device.

“4. The computer-implemented method of claim 1, wherein the first customer data is associated with a customer location during one or both of (i) a date of the financial transaction, and (ii) a time of the financial transaction.

“5. The computer-implemented method of claim 1, wherein the first customer data is associated with customer online shopping activity or online customer online browsing activity.

“6. The computer-implemented method of claim 1, wherein the first customer data is associated with customer shopping preferences or history, customer preferences for types of products or services, or customer-preferred merchants.

“7. The computer-implemented method of claim 1, wherein verifying, by the one or more processors, that the electronic fraud alert is not a false positive based upon analysis of the first customer data includes inputting, by the one or more processors, the first customer data into a second machine learning program that is trained to verify whether electronic fraud alerts are false positives using customer data.

“8. The computer-implemented method of claim 7, wherein the second machine learning program that is trained to verify whether electronic fraud alerts are false positives using customer data uses the first customer data to verify customer location at one or both of (i) a date of the financial transaction, and (ii) a time of the financial transaction.

“9. The computer-implemented method of claim 7, wherein the second machine learning program that is trained to verify whether electronic fraud alerts are false positives using customer data uses the first customer data to verify that customer browsing activity is related to a type of product or service purchased by the financial transaction.

“10. The computer-implemented method of claim 1, wherein one or more facts that caused the electronic fraud alert to be generated include one or more of: a transaction amount; a transaction type; a transaction location or merchant location; a merchant identity; a merchant type; or a number or frequency of transactions.

“11. The computer-implemented method of claim 1, wherein one or more facts that caused the electronic fraud alert to be generated include one or more of: a type of card; a card issuer; a debit or credit card number; a cardholder name; a merchant; a merchant location; a transaction location; a transaction amount; or a type of transaction.

“12. The computer-implemented method of claim 1, wherein verifying that the electronic fraud alert is not a false positive based upon analysis of the customer data includes comparing transaction location data or merchant location data associated with the data detailing the financial transaction to a customer location associated with the customer data to determine a mismatch, and wherein the customer data includes one or both of (i) data from the mobile device of the customer, and (ii) vehicle GPS location data.

“13. The computer-implemented method of claim 1, wherein verifying that the electronic fraud alert is not a false positive based upon analysis of the customer data includes comparing (i) a purchased item, identified using the data detailing the financial transaction, to (ii) previous items purchased by the customer, or items reviewed by the customer as identified by online browsing activity, to identify an unusual or unexpected item purchased.

“14. The computer-implemented method of claim 1, wherein verifying that the electronic fraud alert is not a false positive based upon analysis of the customer data includes comparing a transaction amount identified by the data detailing the financial transaction to a purchasing history of the customer to identify an unusual or unexpected transaction amount or transaction.

“15. A computer-implemented method of detecting whether electronic fraud alerts are false positives, the method comprising: training a machine learning program using fraud classifications made in connection with at least one of a type of transaction data or a value of transaction data associated with a plurality of financial accounts, such that the machine learning program learns a characteristic of the transaction data that is indicative of different fraud classifications; receiving, by one or more processors, transaction data detailing a financial transaction associated with a customer; generating, based at least in part on the data detailing the financial transaction and using a rules-based engine, an electronic fraud alert for the financial transaction; determining, using the trained machine learning program and based at least in part on a type of the transaction data or a value of the transaction data, a reason why the electronic fraud alert was generated; verifying, by the one or more processors, that the electronic fraud alert is not a false positive based upon analysis of: first customer data and the reason; and transmitting, by one or more transceivers, the electronic fraud alert to a mobile device of the customer.

“16. A computer system configured to detect whether electronic fraud alerts are false positives, wherein the computer system comprises one or more processors and one or more transceivers and is configured to: training a machine learning program using fraud classifications made in connection with at least one of a type of transaction data or a value of transaction data associated with a plurality of financial accounts, such that the first machine learning program learns a characteristic of transaction data that is indicative of different fraud classifications; receive transaction data detailing a financial transaction associated with a customer; input the data detailing the financial transaction into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data detailing the financial transaction; when an electronic fraud alert is generated for the financial transaction, input the data detailing the financial transaction into the machine learning program, and determine, using the trained machine learning program and based at least in part on a type of the transaction data or a value of the transaction data, a reason why the electronic fraud alert was generated, determine that the reason why the electronic fraud alert was generated can be verified by customer data; verify that the electronic fraud alert is not a false positive based upon analysis of first customer data; and transmit the electronic fraud alert to a mobile device of the customer.

“17. The computer system of claim 16, wherein the first customer data is associated with a one or both of (i) a date of the financial transaction, and (ii) a time of the financial transaction.

“18. The computer system of claim 16, wherein the first customer data is associated with, or generated and transmitted by, a customer vehicle or a customer mobile device.

“19. The computer system of claim 16, wherein the first customer data is associated with a customer location during one or both of (i) a date of the financial transaction, and (ii) a time of the financial transaction.

“20. The computer system of claim 16, wherein the first customer data is associated with customer online shopping activity or online customer online browsing activity.”

For the URL and additional information on this patent, see: Batra, Reena. Reducing false positives using customer data and machine learning. U.S. Patent Number 11049109, filed June 11, 2020, and published online on June 29, 2021. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11049109.PN.&OS=PN/11049109RS=PN/11049109

(Our reports deliver fact-based news of research and discoveries from around the world.)