Patent Issued for Method and system for correcting errors in consumer reporting (USPTO 11232518): State Farm Mutual Automobile Insurance Company

2022 FEB 10 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- According to news reporting originating from Alexandria, Virginia, by NewsRx journalists, a patent by the inventors Dunn, James J. (Normal, IL, US), Mcwain, Arthur (Bloomington, IL, US), filed on May 8, 2020, was published online on January 25, 2022.

The assignee for this patent, patent number 11232518, is State Farm Mutual Automobile Insurance Company (Bloomington, Illinois, United States).

Reporters obtained the following quote from the background information supplied by the inventors: “Today, consumers’ credit history information is reported to major credit reporting agencies such as Experian, Equifax, TransUnion, and Innovis in a standardized format called Metro 2®. Organizations that report information to these credit reporting agencies may be referred to as herein as “data furnishers.” By providing information in the Metro 2® format, data furnishers meet compliance requirements and are able to report accurate, complete, and timely consumer credit information.

“However, large organizations may have several divisions such as an insurance provider (or financial services provider) that includes several divisions such as an automobile insurance division, a life insurance division, a banking division including a vehicle loan subdivision and a home loan subdivision, an insurance claims division, an insurance sales division, an insurance underwriting division, a homeowner’s insurance division, etc. Members within different divisions or products lines may not have access to consumers’ credit information. As a result, when there is an error on a consumer’s credit report it may be very difficult to access and identify the root cause of the error without access to the consumer’s reported credit information. It may also be difficult to identify recurring inaccuracies in these reports and to proactively correct such inaccuracies throughout the consumer reporting system.”

In addition to obtaining background information on this patent, NewsRx editors also obtained the inventors’ summary information for this patent: “To detect and correct errors in consumer reporting, a consumer reporting system may obtain a secure data file that includes consumer credit information for a consumer, such as a Metro 2® formatted file. The consumer reporting system may then add additional fields to the secure data file which may include a consumer identifier that is non-sensitive private information (non-SPI). The non-SPI consumer identifier may then be used to store the appended secure data file or a portion thereof so that the resulting data file may be searchable and accessible by members within several divisions or products lines throughout an organization, for example, via a client application. Additionally, the non-SPI consumer identifier may be used as a key to retrieve the corresponding data file or portions of the corresponding data file that do not include SPI for display to a user. For example, when a consumer calls the organization and claims that there is an error in her consumer report related to a particular division or product line, a member of the organization in the particular division or product line may retrieve her credit information via the client application by entering her non-SPI consumer identifier. In some instances, the member may compare the consumer’s credit information to user profile information from the consumer’s user profile to identify inconsistencies and determine the cause of the error. For example, the consumer’s user profile may store account information such as bank statements and payment history. This may be compared to the information in the consumer’s credit history to determine whether the consumer made a mistake or interpreted her consumer report incorrectly or whether the consumer report does include errors.

“When there is an error in the consumer report, the user or the consumer reporting system may correct the error by modifying the secure data file. In an implementation, the user or the consumer reporting system may provide the modified secure data file to the credit reporting agencies. In another implementation, a third party vendor may send unmodified information to the credit reporting agencies without interfacing with the user or the consumer reporting system. In some scenarios, the consumer reporting system may automatically identify errors in consumers’ credit information before secure data files are provided to the credit reporting agencies. The consumer reporting system may cross-reference the consumer credit information with user profile information from the consumer’s user profile and may provide an alert to a member of the organization when there is an inconsistency or may replace the consumer credit information with the user profile information from the consumer’s user profile.

“Additionally, the consumer reporting system may identify recurrent errors across several consumers and apply a correction method to current and future secure data files. More specifically, the consumer reporting system may store and analyze previous reporting errors to identify several instances of the same or similar errors within a threshold duration. When the same error occurs with greater than a threshold frequency, the consumer reporting system may apply the same correction method to several current and future secure data files. For example, when an error occurs in the same data field of several secure data files, the consumer reporting system may parse the particular data field in other secure data files to determine whether the particular data field has a same or similar value as the previous secure data files having the error in the particular data field. When the particular data field has a same or similar value as the previous secure data files, the error may be corrected in the same or similar manner as the previous secure data files. In one example, the error may be corrected by applying user profile information for the consumer corresponding to the particular data field. In another example, the error may be corrected by appending an alphanumeric string to the front or back of the value or replacing the value with a particular alphanumeric string. In yet another example, the error may be corrected by applying consumer data from another source that corresponds to the particular data field.

“In this manner, the consumer reporting system provides a tool for reactively and proactively verifying and correcting errors in a consumer’s credit information. Additionally, the present embodiments advantageously identify common errors when secure data files such as Metro 2® formatted files are initially generated. The consumer reporting system may then correct such errors on a system wide level before the secure data files are provided to credit reporting agencies. This may increase the accuracy of consumer credit information and thus, credit reports. Furthermore, the present embodiments advantageously allow for storage and retrieval of consumer credit information outside of a secure data set, so that the consumer credit information is accessible to members across several divisions or products lines of an organization.

“In an embodiment, a computer-implemented method for correcting errors in consumer reporting includes training a machine learning model for identifying errors in consumer credit reporting using (i) a first set of non-sensitive private information (non-SPI) consumer credit information from statements including errors, and (ii) a second set of non-SPI consumer credit information from statements that do not include errors. The method further includes obtaining a secure data file including non-SPI consumer credit information for a consumer related to one or more products and generating a non-sensitive private information (non-SPI) consumer identifier corresponding to the non-SPI consumer credit information, where the non-SPI consumer identifier references the non-SPI consumer credit information. The method also includes receiving, via a client application of a user, a request for non-SPI consumer credit information of the consumer, the request including the non-SPI consumer identifier. In response to verifying that the user has permission to access the non-SPI consumer credit information, the method includes parsing the secure data file to identify one or more particular data fields, and for each particular data field, a subset of the non-SPI consumer credit information corresponding to the particular data field, applying the subset of the non-SPI consumer credit information for each particular data field to the machine learning model to identify an error in the non-SPI consumer credit information, and presenting an indication of the identified error along with a subset of the non-SPI consumer credit information corresponding to the identified error to be displayed via the client application on a client device.”

The claims supplied by the inventors are:

“1. A computer-implemented method for correcting errors in consumer credit reporting, the method executed by one or more processors programmed to perform the method, the method comprising: training, by the one or more processors, a first machine learning model for identifying incorrect value errors in consumer credit reporting using (i) a first set of non-sensitive private information (non-SPI) consumer credit information from statements including incorrect value errors, and (ii) a second set of non-SPI consumer credit information from statements that do not include incorrect value errors; training, by the one or more processors, a second machine learning model for correcting incorrect value errors in consumer credit reporting using original values and changed values for previously corrected errors; obtaining, at the one or more processors, a secure data file including a third set of non-SPI consumer credit information for a consumer related to one or more products; generating, by the one or more processors, a non-SPI consumer identifier corresponding to and referencing the third set of non-SPI consumer credit information; receiving, at the one or more processors from a user, a request for the third set of non-SPI consumer credit information of the consumer, the request including the non-SPI consumer identifier; in response to verifying that the user has permission to access the third set of non-SPI consumer credit information: parsing, by the one or more processors, the secure data file to identify one or more particular data fields, and for each particular data field, a subset of the third set of non-SPI consumer credit information corresponding to each particular data field; applying, by the one or more processors, the subset of the third set of non-SPI consumer credit information for each particular data field to the first machine learning model to identify an incorrect value error in the third set of non-SPI consumer credit information; applying, by the one or more processors, the incorrect value error to the second machine learning model to automatically correct the incorrect value error; and presenting, by the one or more processors, an indication of the identified incorrect value error and a corrected value for the incorrect value error along with the subset of the third set of non-SPI consumer credit information corresponding to the identified incorrect value error to be displayed on a client device.

“2. The method of claim 1, further comprising: retrieving, by the one or more processors, user profile information from a user profile of the consumer; comparing, by the one or more processors, the user profile information to the third set of non-SPI consumer credit information; and identifying, by the one or more processors, an error in the third set of non-SPI consumer credit information based on the comparison.

“3. The method of claim 2, further comprising: correcting, by the one or more processors, the error in the third set of non-SPI consumer credit information by applying the user profile information.

“4. The method of claim 3, further comprising: providing, by the one or more processors, the corrected third set of non-SPI consumer credit information for the consumer to a credit reporting agency.

“5. The method of claim 1, wherein receiving the secure data file including the third set of non-SPI consumer credit information for the consumer includes receiving a plurality of secure data files for a plurality of consumers and further comprising: analyzing, by the one or more processors, a plurality of errors in respective fourth sets of non-SPI consumer credit information of the plurality of consumers to identify a trend in the plurality of errors corresponding to the respective fourth sets of non-SPI consumer credit information.

“6. The method of claim 5, further comprising: receiving, at the one or more processors, another secure data file including a fifth set of non-SPI consumer credit information for another consumer; and in response to identifying the trend, applying, by the one or more processors, a corrective measure to the fifth set of non-SPI consumer credit information for the other consumer to compensate for the trend in the plurality of errors.

“7. The method of claim 1, wherein verifying that the user has permission to access the third set of non-SPI consumer credit information includes: generating, by the one or more processors, a security group having a plurality of users who belong to the security group, wherein each user who belongs to the security group has permission to access one or more sets of non-SPI consumer credit information associated with the security group, the one or more sets of non-SPI consumer credit information including the third set of non-SPI consumer credit information for the consumer; assigning, by the one or more processors, the plurality of users to the security group; and in response to determining that the user is assigned to the security group, determining that the user has permission to access the third set of non-SPI consumer credit information.

“8. A system for correcting errors in consumer reporting, the system comprising: one or more processors; a communication network; a non-transitory computer-readable memory communicatively coupled to the one or more processors and the communication network and storing thereon instructions that, when executed by the one or more processors, cause the system to: train a first machine learning model for identifying incorrect errors in consumer credit reporting using (i) a first set of non-sensitive private information (non-SPI) consumer credit information from statements including incorrect value errors, and (ii) a second set of non-SPI consumer credit information from statements that do not include incorrect value errors; train a second machine learning model for correcting incorrect value errors in consumer credit reporting using original values and changed values for previously corrected errors; obtain a secure data file including a third set of non-SPI consumer credit information for a consumer related to one or more products; generate a non-SPI consumer identifier corresponding to and referencing the third set of non-SPI consumer credit information; receive, from a user, a request for the third set of non-SPI consumer credit information of the consumer, the request including the non-SPI consumer identifier; in response to verifying that the user has permission to access the third set of non-SPI consumer credit information: parse the secure data file to identify one or more particular data fields, and for each particular data field, a subset of the third set of non-SPI consumer credit information corresponding to each particular data field; apply the subset of the third set of non-SPI consumer credit information for each particular data field to the first machine learning model to identify an incorrect value error in the third set of non-SPI consumer credit information; apply the incorrect value error to the second machine learning model to automatically correct the incorrect value error; and present an indication of the identified incorrect value error and a corrected value for the incorrect value error along with the subset of the third set of non-SPI consumer credit information corresponding to the identified incorrect value error to be displayed on a client device.

“9. The system of claim 8, wherein the instructions further cause the system to: retrieve user profile information from a user profile of the consumer; compare the user profile information to the third set of non-SPI consumer credit information; and identify an error in the third set of non-SPI consumer credit information based on the comparison.

“10. The system of claim 9, wherein the instructions further cause the system to correct the error in the third set of non-SPI consumer credit information by applying the user profile information.

“11. The system of claim 10, wherein the instructions further cause the system to provide, via the communication network, the corrected third set of non-SPI consumer credit information for the consumer to a credit reporting agency.

“12. The system of claim 8, wherein to receive the secure data file including the third set of non-SPI consumer credit information for the consumer, the instructions cause the system to receive a plurality of secure data files for a plurality of consumers and the instructions further cause the system to: analyze a plurality of errors in respective fourth sets of non-SPI consumer credit information of the plurality of consumers to identify a trend in the plurality of errors corresponding to the respective fourth sets of non-SPI consumer credit information.

“13. The system of claim 12, wherein the instructions further cause the system to: receive another secure data file including a fifth set of non-SPI consumer credit information for another consumer; and in response to identifying the trend, apply a corrective measure to the fifth set of non-SPI consumer credit information for the other consumer to compensate for the trend in the plurality of errors.

“14. The system of claim 8, wherein to verify that the user has permission to access the third set of non-SPI consumer credit information, the instructions cause the system to: generate a security group having a plurality of users who belong to the security group, wherein each user who belongs to the security group has permission to access one or more sets of non-SPI consumer credit information associated with the security group, the one or more sets of non-SPI consumer credit information including the third set of non-SPI consumer credit information for the consumer; assign the plurality of users to the security group; and in response to determining that the user is assigned to the security group, determine that the user has permission to access the third set of non-SPI consumer credit information.”

There are additional claims. Please visit full patent to read further.

For more information, see this patent: Dunn, James J. Method and system for correcting errors in consumer reporting. U.S. Patent Number 11232518, filed May 8, 2020, and published online on January 25, 2022. 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=11232518.PN.&OS=PN/11232518RS=PN/11232518

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