Patent Issued for Computer-based systems configured to utilize predictive machine learning techniques to define software objects and methods of use thereof (USPTO 11593677): American Express Travel Related Services Company

2023 MAR 21 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- A patent by the inventors Arya, Neha (New York, NY, US), Kong, Xiangzhen (Hoboken, NJ, US), Nagarajan, Venkatesh (Scottsdale, AZ, US), Sharma, Mansi (Edison, NJ, US), filed on December 31, 2019, was published online on February 28, 2023, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents.

Patent number 11593677 is assigned to American Express Travel Related Services Company (New York, New York, United States).

The following quote was obtained by the news editors from the background information supplied by the inventors: “BACKGROUND OF TECHNOLOGY

“Some software objects can have multiple features. Typically, generating software objects can be a technologically daunting task when there are multiple competitive objectives between one or more entities with respect to such software objects.”

In addition to the background information obtained for this patent, NewsRx journalists also obtained the inventors’ summary information for this patent: “SUMMARY OF DESCRIBED SUBJECT MATTER

“In some embodiments, the present disclosure provides technically improved computer-based system/method/apparatus that includes at least the following components/steps of a computer-based prediction system. At least some embodiments are directed to an example of prediction system. The prediction system predicts a first aspect of a user profile utilizing a categorization machine learning model and a user activity profile, the user profile and the user activity profile are associated with a user. The user activity profile comprises a plurality of values associated with demographics and historical data of the user. The prediction system can predict a software object associated with the user profile utilizing an optimization machine learning model, the first aspect of the user profile, and a second aspect of the user profile. The software object is optimized with respect to at least one competitive interest between the user associated with the user profile and an entity associated with the software object. The prediction system outputs the software object to a client computing device of the user.”

The claims supplied by the inventors are:

“1. An apparatus, comprising: a processor; and a non-transitory memory storing instructions which, when executed by the processor, causes the processor to: receive personal identifiable data associated with a user, the personal identifiable data received during a client session from a first computing device; receive a dataset associated with the user based on the personal identifiable data wherein the dataset is received from a second computing device; extract from the dataset, a plurality of values associated with historical data of the user; predict a plurality of software objects customized for the user utilizing a trained categorization machine learning model and the plurality of values associated with the historical data of the user, and for each of the plurality of software objects determine a first aspect from a plurality of features of that software object; perform a risk-assessment analysis using the plurality of values associated with the historical data of the user to generate a set of confidence values, wherein each confidence value from the set of confidence values indicates an association between each software object from the plurality of software objects and a second aspect from the plurality of features of that software object; produce at least one software object utilizing a trained optimization machine learning model with the first aspect of the software object, the second aspect of the software object, and the set of confidence values, wherein the at least one software object is optimized with respect to at least one competitive interest between the user and an entity provider of the at least one software object; and output a notification indicative of the at least one software object to the first computing device associated with the user during the client session.

“2. The apparatus of claim 1, wherein the trained categorization machine learning model comprises a k-nearest neighbor machine learning model, a learning vector quantization machine learning model, or a locally learning machine learning model.

“3. The apparatus of claim 1, wherein the plurality of features of the software object comprise a credit line, an annual percentage rate, a type of rewards program, a membership program, or a health insurance program.

“4. The apparatus of claim 1, wherein the at least one competitive interest comprises maximizing a profit of the entity provider and minimizing an interest rate associated with the software object, or maximizing a probability that the user accepts the at least one software object and minimizing a risk of monetary loss by the entity associated with the at least one software object.

“5. The apparatus of claim 1, wherein the trained optimization machine learning model comprises a gradient boosting machine learning model, a random forest model, a bootstrap aggregation model, a stacked generalization model, a gradient boosted regression tree model, or a radial basis function network model.

“6. The apparatus of claim 1, wherein the processor is further configured to: retrain the trained categorization machine learning model and the trained optimization machine learning model using the dataset associated with the user when the user declines the at least one software object.

“7. The apparatus of claim 1, wherein the trained optimization machine learning model is a gradient boosting machine learning model and the gradient boosting machine learning model includes multiple decision tree models.

“8. A method, comprising: receiving, by a processor, personal identifiable data associated with a user, the personal identifiable data received during a client session from a first computing device; receiving, by the processor, a dataset associated with the user based on the personal identifiable data wherein the dataset is received from a second computing device; extracting, by the processor, a plurality of values associated with historical data of the user; predicting, by the processor, a plurality of software objects customized for the user utilizing a trained categorization machine learning model and the plurality of values associated with the historical data of the user, and for each of the plurality of software objects determine a first aspect from a plurality of features of that software object; performing, by the processor, a risk-assessment analysis using the plurality of values associated with the historical data of the user to generate a set of confidence values, wherein each confidence value from the set of confidence values indicates an association between each software object from the plurality of software objects and a second aspect from the plurality of features of that software object; producing, by the processor, at least one software object utilizing a trained optimization machine learning model with the first aspect of the software object, the second aspect of the software object, and the set of confidence values, wherein the at least one software object is optimized with respect to at least one competitive interest between the user and an entity provider of the at least one software object; and outputting, by the processor, a notification indicative of the at least one software object to the first computing device associated with the user during the client session.

“9. The method of claim 8, wherein the trained categorization machine learning model comprises a k-nearest neighbor machine learning model, a learning vector quantization machine learning model, or a locally learning machine learning model.

“10. The method of claim 8, wherein the plurality of features of the software object comprise a credit line, an annual percentage rate, a type of rewards program, a membership program, or a health insurance program.

“11. The method of claim 8, wherein the at least one competitive interest comprises maximizing a profit of the entity provider and minimizing an interest rate associated with the software object, or maximizing a probability that the user accepts the at least one software object and minimizing a risk of monetary loss by the entity associated with the at least one software object.

“12. The method of claim 8, wherein the trained optimization machine learning model comprises a gradient boosting machine learning model, a random forest model, a bootstrap aggregation model, a stacked generalization model, a gradient boosted regression tree model, or a radial basis function network model.

“13. The method of claim 8, further comprising: retraining the trained categorization machine learning model and the trained optimization machine learning model using the dataset associated with the user when the user declines the at least one software object.

“14. The method of claim 8, wherein the trained optimization machine learning model is a gradient boosting machine learning model and the gradient boosting machine learning model includes multiple decision tree models.

“15. A non-transitory computer readable medium comprising code which, when executed by a processor, causes the processor to: receive personal identifiable data associated with a user, the personal identifiable data received during a client session from a first computing device; receive a dataset associated with the user based on the personal identifiable data wherein the dataset is received from a second computing device; extract from the dataset, a plurality of values associated with historical data of the user; predict a plurality of software objects customized for the user utilizing a trained categorization machine learning model and the plurality of values associated with the historical data of the user, and for each of the plurality of software objects determine a first aspect from a plurality of features of that software object; perform a risk-assessment analysis using the plurality of values associated with the historical data of the user to generate a set of confidence values, wherein each confidence value from the set of confidence values indicates an association between each software object from the plurality of software objects and a second aspect from the plurality of features of that software object; produce at least one software object utilizing a trained optimization machine learning model with the first aspect of the software object, the second aspect of the software object, and the set of confidence values, wherein the at least one software object is optimized with respect to at least one competitive interest between the user and an entity provider of the at least one software object; and output a notification indicative of the at least one software object to the first device associated with the user during the client session.

“16. The non-transitory computer readable medium of claim 15, wherein the trained categorization machine learning model comprises a k-nearest neighbor machine learning model, a learning vector quantization machine learning model, or a locally learning machine learning model.

“17. The non-transitory computer readable medium of claim 15, wherein the plurality of features of the software object comprise a credit line, an annual percentage rate, a type of rewards program, a membership program, or a health insurance program.

“18. The non-transitory computer readable medium of claim 15, wherein the at least one competitive interest comprises maximizing a profit of the entity provider and minimizing an interest rate associated with the software object, or maximizing a probability that the user accepts the at least one software object and minimizing a risk of monetary loss by the entity associated with the at least one software object.”

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

URL and more information on this patent, see: Arya, Neha. Computer-based systems configured to utilize predictive machine learning techniques to define software objects and methods of use thereof. U.S. Patent Number 11593677, filed December 31, 2019, and published online on February 28, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11593677)&db=USPAT&type=ids

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