US 12,242,942 B2
Utilizing machine learning models to identify insights in a document
Joni Bridget Jezewski, Dublin, CA (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Dec. 19, 2023, as Appl. No. 18/544,957.
Application 18/544,957 is a continuation of application No. 17/647,187, filed on Jan. 6, 2022, granted, now 11,861,477.
Application 17/647,187 is a continuation of application No. 16/693,650, filed on Nov. 25, 2019, granted, now 11,227,121, issued on Jan. 8, 2022.
Application 16/693,650 is a continuation of application No. 16/385,496, filed on Apr. 16, 2019, granted, now 10,489,512, issued on Nov. 26, 2019.
Application 16/385,496 is a continuation of application No. 15/896,922, filed on Feb. 14, 2018, granted, now 10,303,771, issued on May 28, 2019.
Prior Publication US 2024/0127126 A1, Apr. 18, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/58 (2019.01); G06F 16/30 (2019.01); G06F 18/24 (2023.01); G06F 18/2413 (2023.01); G06F 40/216 (2020.01); G06F 40/284 (2020.01); G06F 40/30 (2020.01); G06N 3/04 (2023.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); G06V 30/40 (2022.01)
CPC G06N 20/20 (2019.01) [G06F 16/30 (2019.01); G06F 18/24 (2023.01); G06F 18/2413 (2023.01); G06F 40/216 (2020.01); G06F 40/284 (2020.01); G06F 40/30 (2020.01); G06N 3/04 (2013.01); G06N 20/00 (2019.01); G06V 30/40 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
providing, by a device, a user interface that displays document information;
generating, by the device and based on a mechanism on the user interface being selected, a request to identify insights in the document information;
providing, by the device and based on the mechanism being selected, the document information and the request to an insight platform,
wherein information indicating an abstract insight, a first concrete insight, and a non-insight is identified in the document information using a machine learning model based on the request,
wherein the machine learning model is trained based on historical information provided to the machine learning model, prediction information based on the historical information provided to the machine learning model, and updates to the machine learning model based on the prediction information,
wherein the abstract insight is matched with a second concrete insight that is different than the first concrete insight and the abstract insight using the machine learning model, and
wherein the non-insight is matched with a stored insight that is different from insights identified in the document information using the machine learning model; and
providing, by the device and for display, an insight document that is generated based on a combination of the first concrete insight, the abstract insight matched with the second concrete insight, and the stored insight,
wherein the first concrete insight, the abstract insight matched with the second concrete insight, and the stored insight are grouped together in the insight document.