US 12,340,330 B2
Systems and methods for recommending tasks for execution by third party services
Yoky Matsuoka, Los Altos Hills, CA (US); Nitin Viswanathan, San Francisco, CA (US); Gwendolyn W. van der Linden, Redwood City, CA (US); Malia Beaulieu, San Jose, CA (US); Lingyun Liu, Sunnyvale, CA (US); Benjamin Deming, Campbell, CA (US); and Sean Paterson, Mountain View, CA (US)
Assigned to Yohana LLC, Palo Alto, CA (US)
Filed by Yohana LLC, Palo Alto, CA (US)
Filed on Aug. 19, 2022, as Appl. No. 17/820,915.
Claims priority of provisional application 63/234,856, filed on Aug. 19, 2021.
Prior Publication US 2023/0057896 A1, Feb. 23, 2023
Int. Cl. G06Q 10/00 (2023.01); G06Q 10/0631 (2023.01)
CPC G06Q 10/06316 (2013.01) 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving sensor data associated with a user, wherein the sensor data is stored in association with a user model that corresponds to the user;
generating a feature vector from the sensor data, a first training dataset, and the user model;
executing a machine-learning model using the feature vector, wherein the machine-learning model generates a recommendation for each of one or more tasks;
receiving authorization to execute a particular task;
generating one or more proposals, wherein proposals include a different set of one or more procedures from other proposals that when executed, execute the particular task;
receiving a selection of a particular proposal of the one or more proposals;
facilitating a transmission that includes an identification of the particular proposal;
generating feedback using the recommendation and the identification of the particular proposal, wherein the feedback includes an evaluation of the recommendation and the particular proposal;
generating a second training dataset associated with the machine-learning model by appending the feedback to the first training dataset, wherein generating the second training dataset causes a modification in subsequent iterations of the machine-learning model to improve task prediction for recommendations;
executing the machine-learning model using the second training dataset to generate a new one or more tasks, wherein the machine-learning model is executed in response to generating the second training dataset; and
facilitating a display of the new one or more tasks generated using the second training dataset.