US 12,217,276 B2
Attribute selection for matchmaking
Krishna Kolli, Dublin, CA (US)
Assigned to Intuit Inc., Mountain View, CA (US)
Filed by Intuit Inc., Mountain View, CA (US)
Filed on Sep. 7, 2023, as Appl. No. 18/243,368.
Application 18/243,368 is a continuation of application No. 16/059,482, filed on Aug. 9, 2018, granted, now 11,790,381.
Prior Publication US 2023/0419344 A1, Dec. 28, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/02 (2023.01); G06F 18/214 (2023.01); G06Q 30/0202 (2023.01)
CPC G06Q 30/0202 (2013.01) [G06F 18/214 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method for assisting users with improving the effectiveness of their profiles, the method performed by one or more processors of a system and comprising:
storing profile data in a user profile database of the system, the profile data including, for each respective user of a plurality of users associated with the system, a profile identifying attributes associated with the respective user;
obtaining, from the user profile database using the one or more processors, a selection data vector including values each indicating a selection rate for a respective user, each selection rate indicative of a rate at which users of the system have chosen to receive services from the respective user after selecting the respective user's profile;
generating, for a selected user associated with a user computing environment used to interact with the system, selection prediction data for the selected user by analyzing, using an analysis model, the profile data associated with the selected user, the selection prediction data predicting, for each respective change of a set of possible changes to the selected user's profile, how the selection rate for the selected user will change if the selected user's profile is adjusted in accordance with the respective change, wherein the analysis model is trained, using a model training module in conjunction with a machine learning process, to predict how changes in a profile of a given user will affect the selection rate for the given user based on feeding the profile data associated with the given user to the analysis model and comparing the values included in the selection data vector with selection rates included in a selection rate vector generated for the plurality of users based on the attributes, and wherein the analysis model is retrained, using the model training module in conjunction with the machine learning process, until a difference between the values included in the selection data vector and the values included in the selection rate vector is below a threshold;
selecting, from the selection prediction data, one or more recommended changes likely to result in an increase in the selection rate for the selected user; and
outputting, to the selected user via an interface module communicatively coupled to the user computing environment, recommended profile adjustment data including a prompt recommending that the selected user make at least one of the one or more recommended changes to the selected user's profile.