US 12,271,742 B2
Automatically executing application routines with user inputs
Cliff Kuang, San Francisco, CA (US); Diana Avram, Zurich (CH); Mugurel-Ionut Andreica, Adliswil (CH); Radu Voroneanu, Zurich (CH); Sneha Ashok, San Jose, CA (US); Deepak Goyal, Sunnyvale, CA (US); Kyunghoon Lee, Santa Clara, CA (US); Alice Liang, Sunnyvale, CA (US); Dana Ritter, Horgen (CH); Adam Coimbra, Los Altos, CA (US); Anton Berezin, Zurich (CH); and Andre Elisseeff, Basel (CH)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by GOOGLE LLC, Mountain View, CA (US)
Filed on Aug. 4, 2023, as Appl. No. 18/230,561.
Prior Publication US 2025/0045071 A1, Feb. 6, 2025
Int. Cl. G06F 9/451 (2018.01); G06F 3/0482 (2013.01); G06F 3/0484 (2022.01)
CPC G06F 9/451 (2018.02) [G06F 3/0482 (2013.01); G06F 3/0484 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method implemented by one or more processors, the method comprising:
identifying a plurality of instances of a user performing an action, wherein the action is performed via an application, and wherein the action is at least a portion of a sequence of actions;
determining, based on the instances, one or more user interactions initiated by the user while performing the action, wherein a given user interaction indicates a specified input of the user while performing one or more of the actions;
generating, based on a quantity of each specified input for a given user interaction, a confidence metric;
selecting, based on the confidence metrics, a corresponding rendering type for the application while performing the action, wherein the corresponding rendering type indicates a format to utilize while rendering the application via one or more output devices;
storing the sequence of actions with an association to the corresponding rendering type;
receiving a request to execute the sequence of actions; and
rendering the application in accordance with the rendering type.