| CPC G06N 20/00 (2019.01) [G06F 30/20 (2020.01); G06N 5/046 (2013.01); G06Q 30/02 (2013.01)] | 20 Claims |

|
1. A computer-implemented method comprising:
receiving a request to access a web page of a website associated with a content provider;
accessing a set of candidate design attributes for the web page;
generating a set of web-page permutations associated with the web page, wherein a web-page permutation defines visual characteristics of web content presented on the web page, and wherein the visual characteristics are associated with a combination of one or more candidate design attributes;
applying a trained model to the set of web-page permutations to generate a set of predictive values, wherein a predictive value identifies a probability of a user device interacting with the web content defined by a web-page permutation associated with a particular combination of candidate design attributes;
selecting, based on the set of predictive values, an optimized web-page permutation associated with the web page, wherein the optimized web-page permutation defines visual characteristics that are associated with a subset of the set of candidate design attributes;
modifying the web page using the optimized web-page permutation, wherein the modified web page is configured to present the web content having the visual characteristics associated with the subset of candidate design attributes; and
facilitating display of the modified web page on the user device.
|