US 10,891,351 B2
Systems and methods for automatically generating and optimizing web pages
Vincent Pham, Champaign, IL (US)
Assigned to CAPITAL ONE SERVICES, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Nov. 29, 2018, as Appl. No. 16/204,054.
Prior Publication US 2020/0175088 A1, Jun. 4, 2020
Int. Cl. G06F 17/00 (2019.01); G06F 16/957 (2019.01); G06F 11/34 (2006.01); G06F 16/958 (2019.01); G06F 11/36 (2006.01)
CPC G06F 16/9577 (2019.01) [G06F 11/3466 (2013.01); G06F 16/958 (2019.01); G06F 16/9574 (2019.01); G06F 11/368 (2013.01); G06F 11/3672 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for generating and selecting web page layouts based on performance analytics, the method comprising:
receiving, by a first device, code associated with a default web page;
parsing, by a processor associated with the first device, the code associated with the default web page by:
identifying one or more components of the default web page;
identifying attributes of the one or more components of the default web page; and
associating the one or more components with at least one pre-set component type and at least one pre-set attribute type;
receiving, from a user of the first device, a plurality of user web page preferences, wherein the plurality of user web page preferences comprise at least two factors selected from average user engagement time, unique views, click-through rate, and product purchase rate;
retrieving, by the first device, historical web site performance data, wherein the historical web site performance data represents performance data for a plurality of previous web page versions viewed by a plurality of web page visitors, each of the plurality of web page visitors associated with at least one second device of a plurality of second devices;
assigning, by the processor, an associated weight for each of the historical web site performance data, the plurality of user web page preferences, and the parsed code associated with the default web page;
generating, by the processor, based on the plurality of user web page preferences, the historical web site performance data, the parsed code associated with the default web page, and the associated weight for each of the plurality of user web page preferences, the historical web site performance data, and the parsed code associated with the default web page, a plurality of web page versions;
transmitting, by the user of the first device, the plurality of web page versions to the plurality of second devices associated with the plurality of web page visitors;
collecting performance data associated with each of the plurality of web page versions;
analyzing, by the processor, the performance data for the plurality of web page versions by executing a multi-arm bandit algorithm for each web page version of the plurality of web page versions;
storing the performance data for at least one of the plurality of web page versions amongst the historical web site performance data; and
based on the performance data and a rank of the plurality of user web page preferences, selecting one of the plurality of web page versions as the default web page.