IP Library › Granted Patent US 12,614,216
Granted Patent B2
US 12,614,216 · App. 18/432,293 · Granted Apr 28, 2026

Preemptive transaction analysis

Inventors: Michael Mossoba (Great Falls, VA); Joshua Edwards (Philadelphia, PA); Abdelkadar M'Hamed Benkreira (Washington, DC)
Assignee: Capital One Services, LLC
G06Q30/0601G06Q20/40G06Q20/401G06Q20/4016G06Q30/0623G06Q30/0624G06Q30/0625G06Q30/0633G06Q30/0637H04L63/1408
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,614,216
App. No.
18/432,293
Granted
Apr 28, 2026
Kind
B2
Abstract

A method may include receiving activity data associated with a user, wherein the activity data relates to online activity involving a product type, identifying the product type associated with the activity data, and predicting, based on the activity data, that the user is likely to purchase a product of the product type. The method may include generating, based on predicting that the user is likely interested in purchasing the product of the product type, an annotation to indicate that a potential transaction to purchase the product is forthcoming, and storing the annotation in a profile associated with an account of the user. The method may include detecting a transaction to purchase the product, wherein the transaction involves a payment from the account, and performing an action associated with a fraud analysis of the transaction based on the annotation.

Claims (89)

1 . A method, comprising:

monitoring, by a device, activity performed on the device based on an application running on the device,

wherein one or more images are associated with the monitored activity;

generating, by the device and based on processing the monitored activity using the one or more images, activity data,

wherein processing the one or more images comprises analyzing the one or more images using at least one of:

a computer image processing technique that includes at least one of object detection from the one or more images, or edge detection of the one or more images,

a text processing technique, or

a code processing technique to recognize one or more products;

predicting, by the device and based on using a machine learning model, that a user is likely to purchase a particular type of product,

wherein the machine learning model is trained to predict a likelihood of the user purchasing the particular type of product based on the activity data;

generating, by the device and based on predicting that the user is likely to purchase the particular type of product, a flag associated with a potential transaction associated with purchasing the particular type of product;

storing the flag with a profile of the user;

detecting, by the device and within a threshold time period after the generation of the flag, a transaction associated with a purchase of a product related to the particular type of product;

selectively bypassing, by the device and based on the flag, a subset of processes from one or more fraud analysis processes to be performed by the device in relation to the transaction,

wherein the flag is configured to expire after the threshold time period passing without a related transaction being performed, and

wherein the flag is removed or deleted from storage upon expiration;

performing, by the device and based on the bypassing, a subset of the one or more fraud analysis processes; and

sending, by the device and based on performing the particular set of processes, an authorization associated with the transaction.

2 . The method of claim 1 , further comprising:

analyzing the activity data using an image processing technique; and

determining that the activity data is associated with the particular type of product.

3 . The method of claim 1 , wherein the threshold time period is determined based on historical information associated with the activity data.

4 . The method of claim 1 , wherein the data is received from another device.

5 . The method of claim 1 , wherein performing the particular set of processes comprises:

obtaining account information associated with the user; and

authorizing the transaction based on the account information.

6 . The method of claim 1 , wherein the bypassing comprises:

bypassing an examination process based on the flag.

7 . The method of claim 1 , wherein the activity is related to online activity comprising capturing an image of an item, sending a message identifying the product, accessing offline media associated with the item, purchasing related items, traveling to a location of a merchant that sells the item, or combinations thereof.

8 . A device, comprising:

one or more memories; and

one or more processors, coupled to the one or more memories, configured to:

monitor, by the device, activity performed on the device based on an application running on the device,

wherein one or more images are associated with the monitored activity;

generate, based on processing the monitored activity using the one or more images, activity data,

wherein processing the one or more images comprises analyzing the one or more images using at least one of:

a computer image processing technique that includes at least one of object detection from the one or more images, or edge detection of the one or more images,

a text processing technique, or

a code processing technique to recognize one or more products;

generate, based on predicting that a user is likely to purchase a particular type of product, a flag associated with a potential transaction associated with purchasing the particular type of product and storing the flag with a profile of the user,

wherein the prediction is based on using a machine learning model that is trained to predict a likelihood of the user in purchasing the particular type of product based on the activity data;

detect, within a threshold time period after the generation of the flag, a transaction associated with a purchase of a product related to the particular type of product;

selectively bypass, based on the flag, a subset of processes from one or more fraud analysis processes to be performed by the device in relation to the transaction,

wherein the flag is configured to expire after the threshold time period passing without a related transaction being performed, and

wherein the flag is removed or deleted from storage upon expiration;

perform, based on the bypassing, a particular set of processes from one or more fraud analysis processes associated with the transaction; and

send, based on performing the particular set of processes, an authorization associated with the transaction.

9 . The device of claim 8 , wherein the one or more processors are further configured to:

analyze the activity data using an image processing technique; and

determine that the activity data is associated with the particular type of product.

10 . The device of claim 8 , wherein the threshold time period is determined based on historical information associated with the activity data.

11 . The device of claim 8 , wherein the one or more processors, to perform the particular set of processes, are configured to:

obtain account information associated with the user; and

authorize the transaction based on the account information.

12 . The device of claim 8 , wherein the one or more processors, when bypassing, are further configured to:

bypass an examination process based on the flag.

13 . The device of claim 9 , wherein the activity is related to online activity comprising capturing an image of a product, sending a message identifying the product, accessing offline media associated with the product, purchasing related products, traveling to a location of a merchant that sells the product, or combinations thereof.

14 . The device of claim 8 , wherein the monitored activity is related to at least one of:

online activity,

the one or more images that is captured by one or more camera devices,

information obtained via a radio frequency identification (RFID) device, or

location information obtained via a global positioning system (GPS) device.

15 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:

one or more instructions that, when executed by one or more processors of a device, cause the device to:

monitor, by the device, activity performed on the device based on an application running on the device,

wherein one or more images are associated with the monitored activity;

generate, based on processing the monitored activity using the one or more images, activity data,

wherein processing the one or more images comprises analyzing the one or more images using at least one of:

a computer image processing technique that includes at least one of object detection from the one or more images, or edge detection of the one or more images,

a text processing technique, or

a code processing technique to recognize one or more products;

generate, based on predicting that a user is likely to purchase a particular type of product, a flag associated with a potential transaction associated with purchasing the particular type of product and storing the flag with a profile of the user,

wherein the prediction is based on using a machine learning model that is trained to predict a likelihood of the user in purchasing the particular type of product based on the activity data;

detect, within a threshold time period after the generation of the flag, a transaction associated with a purchase of a product related to the particular type of product;

selectively bypass, based on the flag, a subset of processes from one or more fraud analysis processes to be performed by the device in relation to the transaction,

wherein the flag is configured to expire after the threshold time period passing without a related transaction being performed, and

wherein the flag is removed or deleted from storage upon expiration;

perform, based on the bypassing, a set of processes from one or more fraud analysis processes associated with the transaction; and

send, based on performing the particular set of processes, an authorization associated with the transaction.

16 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions further cause the device to:

analyze the activity data using an image processing technique; and

determine that the activity data is associated with the particular type of product.

17 . The non-transitory computer-readable medium of claim 15 , wherein the threshold time period is determined based on historical information associated with the activity data.

18 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to perform the particular set of processes, cause the device to:

obtain account information associated with the user; and

authorize the transaction based on the account information.

19 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to bypass, further cause the device to:

bypass an examination process based on the flag.

20 . The non-transitory computer-readable medium of claim 15 , wherein the activity is related to online activity comprising capturing an image of an item, sending a message identifying the product, accessing offline media associated with the item, purchasing related items, traveling to a location of a merchant that sells the item, or combinations thereof.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 5, 2024
From: MOSSOBA, MICHAEL; EDWARDS, JOSHUA; BENKREIRA, ABDELKADAR M'HAMED
To: CAPITAL ONE SERVICES, LLC
Reel/Frame 066346/0878 →
Continuity (4)
Continuation 17664438 · May 23, 2022
Continuation 16729949 · Dec 30, 2019
Continuation 16241688 · Jan 7, 2019
Related Publication 20240185315A1 · Jun 6, 2024
References Cited (57)
US 6278481B1 · Schmidt · 2001 [cited by examiner]
US 7324964B2 · Gronberg et al. · 2008 [cited by applicant]
US 7398925B2 · Tidwell et al. · 2008 [cited by applicant]
US 7769638B1 · Mollett et al. · 2010 [cited by applicant]
US 8032449B2 · Hu et al. · 2011 [cited by applicant]
US 8146156B2 · King et al. · 2012 [cited by applicant]
US 8271394B1 · Bogaard · 2012 [cited by applicant]
US 8751486B1 · Neeman et al. · 2014 [cited by applicant]
US 9836733B2 · Course et al. · 2017 [cited by applicant]
US 10134040B2 · Ding et al. · 2018 [cited by applicant]
US 10354184B1 · Vitaladevuni et al. · 2019 [cited by applicant]
US 10460382B2 · Dominguez · 2019 [cited by applicant]
US 10521837B1 · Mossoba · 2019 [cited by applicant]
US 11341548B2 · Mossoba et al. · 2022 [cited by applicant]
US 11488400B2 · Zucker · 2022 [cited by examiner]
US 20020123937A1 · Pickover et al. · 2002 [cited by applicant]
US 20030069820A1 · Hillmer et al. · 2003 [cited by applicant]
US 20030070080A1 · Rosen · 2003 [cited by examiner]
US 20070005437A1 · Stoppelman · 2007 [cited by applicant]
US 20070050840A1 · Grandcolas · 2007 [cited by examiner]
US 20070119918A1 · Hogg et al. · 2007 [cited by applicant]
US 20070198361A1 · Ronning et al. · 2007 [cited by applicant]
US 20080114829A1 · Button · 2008 [cited by examiner]
US 20090216571A1 · Sunshine et al. · 2009 [cited by applicant]
US 20090265211A1 · May et al. · 2009 [cited by applicant]
US 20100114654A1 · Lukose et al. · 2010 [cited by applicant]
US 20110238510A1 · Rowen et al. · 2011 [cited by applicant]
US 20110239158A1 · Barraclough · 2011 [cited by examiner]
US 20120109821A1 · Barbour · 2012 [cited by examiner]
US 20140006218A1 · Muthu · 2014 [cited by applicant]
US 20140114743A1 · Fano · 2014 [cited by examiner]
US 20140279280A1 · Bennett et al. · 2014 [cited by applicant]
US 20150032625A1 · Dill et al. · 2015 [cited by applicant]
US 20150127489A1 · Vasthimal et al. · 2015 [cited by applicant]
US 20150134518A1 · Turovsky et al. · 2015 [cited by applicant]
US 20150382195A1 · Grim et al. · 2015 [cited by applicant]
US 20160014715A1 · Patil et al. · 2016 [cited by applicant]
US 20160078471A1 · Hamedi · 2016 [cited by applicant]
US 20170270544A1 · Jaidka et al. · 2017 [cited by applicant]
US 20170364967A1 · Parikh et al. · 2017 [cited by applicant]
US 20180096372A1 · Rickard, Jr. · 2018 [cited by examiner]
US 20180174222A1 · Venkatakrishnan et al. · 2018 [cited by applicant]
US 20180278552A1 · Quock et al. · 2018 [cited by applicant]
US 20180285838A1 · Franaszek et al. · 2018 [cited by applicant]
US 20180300788A1 · Mattingly et al. · 2018 [cited by applicant]
US 20190012683A1 · Jang et al. · 2019 [cited by applicant]
US 20190043056A1 · Cowan · 2019 [cited by applicant]
US 20220284490A1 · Mossoba et al. · 2022 [cited by applicant]
CN 102812488A · 2012 [cited by applicant]
CN 108492138A · 2018 [cited by applicant]
IN 201500234I3 · 2016 [cited by applicant]
WO WO03054625A1 · 2003 [cited by examiner]
WO WO2017116769A1 · 2017 [cited by examiner]
Sagar P. Jumde, et al ; “Fraud Detection with the Help of Hidden Markov Model and Neural Network”; “National Level Conference” On Mar. 9, 2013 , Yavatmal [MS] India retrieved from Dialog on Sep. 21, 2024 (Year: 2013). [cited by examiner]
Article, “Sober Q4 results expected to set tone for power sector in 2014”; Testa, Dan. SNL Energy Finance Daily [Charlottesville] Jan. 27, 2014, retrieved from Dialog on Sep. 21, 2024. (Year: 2014). [cited by examiner]
N. Ramesh and T.-S. Moh, “Outfit Recommender System,” 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), Barcelona, Spain, 2018, pp. 903-910, retrieved from IP. Com on De… [cited by examiner]
Bouhnik V., et al., “Behavioral Analysis: The Future of Fraud Prevention,” Secured Touch, Dec. 27, 2015, pp. 1-12. [cited by applicant]