US 12,288,144 B2
Machine learning system to predict causal treatment effects of actions performed on websites or applications
Scott Kramer, New York City, NY (US); Cynthia Rogers, Palo Alto, CA (US); Eric Pollmann, Los Altos, CA (US); and Muhammad Bilal Mahmood, San Francisco, CA (US)
Assigned to AMPLITUDE, INC., San Francisco, CA (US)
Filed by AMPLITUDE, INC., San Francisco, CA (US)
Filed on Apr. 12, 2024, as Appl. No. 18/633,929.
Application 18/633,929 is a continuation of application No. 17/967,617, filed on Oct. 17, 2022, granted, now 11,960,980.
Application 17/967,617 is a continuation of application No. 16/525,457, filed on Jul. 29, 2019, granted, now 11,475,357, issued on Oct. 18, 2022.
Prior Publication US 2024/0370772 A1, Nov. 7, 2024
Int. Cl. G06N 20/00 (2019.01); G06F 9/451 (2018.01); G06F 11/34 (2006.01); G06F 17/16 (2006.01); G06F 17/18 (2006.01); G06Q 30/02 (2023.01); H04L 67/50 (2022.01)
CPC G06N 20/00 (2019.01) [G06F 9/451 (2018.02); G06F 11/3438 (2013.01); G06F 11/3466 (2013.01); G06F 17/16 (2013.01); G06F 17/18 (2013.01); G06Q 30/02 (2013.01); H04L 67/535 (2022.05)] 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
one or more processors; and
a non-transitory memory storing instructions that, when executed by the one or more processors, cause performance of:
receiving, for a user interface, interaction data for a period of time that identifies a plurality of users and a plurality of actions performed by the plurality of users through the user interface during the period of time;
identifying a target action, from the plurality of actions, as an output variable;
generating a feature matrix comprising matrix cell values providing an indication of user performance or non-performance of the plurality of actions;
training a machine learning model using the feature matrix as input and values providing an indication of user performance or non-performance of the target action as output;
identifying a treatment action from the plurality of actions;
computing, using the machine learning model, a treatment effect for the identified treatment action, the treatment effect providing an indication of an impact that performance of the treatment action has on eventual performance of the target action; and
facilitating an update to the user interface based at least in part on the computed treatment effect.