US 12,282,862 B2
Method and system for providing a graphical user interface using machine learning and movement of the user or user device
Daniel Ma, Toronto (CA); and Gregory Jason Fields, Waterloo (CA)
Assigned to MYPLANET INTERNET SOLUTIONS LTD., Toronto (CA)
Filed by MYPLANET INTERNET SOLUTIONS LTD, Toronto (CA)
Filed on Feb. 4, 2021, as Appl. No. 17/167,790.
Claims priority of provisional application 62/971,438, filed on Feb. 7, 2020.
Claims priority of provisional application 62/971,420, filed on Feb. 7, 2020.
Prior Publication US 2021/0248490 A1, Aug. 12, 2021
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01); G06F 3/0482 (2013.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01); G06F 3/0482 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method, comprising:
training a context analyzer machine learning model executing on one ore more computers with training motion data for a training activity context, the context analyzer for predicting a predicted activity context from new motion data, wherein the predicted activity context characterizes a particular kinetic state of a user;
training a response analyzer machine learning model executing on the one or more computers with training usage data and training user responses for the training activity context for a user interface, the response analyzer for predicting a predicted user response from the predicted activity context and new usage data;
inputting motion data received from sensors proximate to a user that characterizes an angular velocity and a linear acceleration of the user into the context analyzer to determine an activity context of the user;
inputting the activity context of the user into the response analyzer to determine a predicted user response;
inputting the predicted user response and the activity context of the user into a machine learning system executing on the one or more computers and the machine learning system selects a user interface from a plurality of user interfaces based on the predicted user response, wherein:
a first user interface of the plurality of user interfaces has call to action (CTA) components for a set of user intentions for executing operations of a particular software application;
a second user interface of the plurality of user interfaces has CTA components for a first subset of user intensions selected from the set of user intentions for executing operations of the particular software application, the first subset of user intentions excluding at least one user intention in the set of user intentions; and
a third user interface of the plurality of user interface has CTA components for a second subset of the set of user intentions for executing operations of the particular software application, the second subset of user intentions excluding at least one user intention in the set of user intentions, and the second subset of user intentions being different than the first subset of user intentions, wherein:
the machine learning system selects the second user interface of the plurality of user interfaces in response to the activity context of the user indicating that the user is experiencing a bumpy motion, and the user has a pace that meets a first threshold; and
the machine learning system selects the third user interface of the plurality of user interfaces in response to the activity context of the user indicating that the user has a pace that meets a second threshold that is greater than the first threshold; and
outputting the selected user interface for the user on a user device.