US 12,136,117 B2
Multi-layer optimization for a multi-sided network service
Yuyan Wang, San Francisco, CA (US); Xian Xing Zhang, Mountain View, CA (US); Isaac Suyu Liu, Fremont, CA (US); Yuanchi Ning, San Francisco, CA (US); and Chen Peng, Sunnyvale, CA (US)
Assigned to Uber Technologies, Inc., San Francisco, CA (US)
Filed by Uber Technologies, Inc., San Francisco, CA (US)
Filed on Aug. 27, 2021, as Appl. No. 17/459,708.
Application 17/459,708 is a division of application No. 16/116,054, filed on Aug. 29, 2018, granted, now 11,127,066.
Prior Publication US 2021/0390610 A1, Dec. 16, 2021
Int. Cl. G06Q 30/00 (2023.01); G06F 16/2457 (2019.01); G06F 16/248 (2019.01); G06F 16/9535 (2019.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G06Q 30/0601 (2023.01)
CPC G06Q 30/0631 (2013.01) [G06F 16/24578 (2019.01); G06F 16/248 (2019.01); G06F 16/9535 (2019.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method for recommending entities to users on a platform for a network service executing on a network server, the computer-implemented method comprising:
obtaining, from a client device at a demand prediction module executing on a processor of a recommendation system for the network service on the network server, user features related to a user, the client device executing an application program interface configured to allow the client device to interact with the network service over a communication network;
obtaining, at the demand prediction module, entity features related to an entity associated with the network service;
obtaining, at the demand prediction module, current contextual features;
generating, using the demand prediction module, a likelihood score quantifying whether the user will find the entity to be favorable, the likelihood score based on the user features, entity features, and contextual features;
generating, using an optimization module executing on the processor of the recommendation system, a set of objective values related to a set of objectives for the network service on the network server, each objective value of the set of objective values generated by a different trained computer model, wherein:
at least one of the objective values is a consumer conversion rate that represents a rate of users who act on a recommendation including the entity compared with a rate of users who act on a recommendation including other entities associated with the network service, and
at least one of the objective values is a marketplace fairness score that represents how frequently the entity receives exposure to client devices of users compared with other entities associated with the network service;
generating, using the optimization module, a user recommendation score for the entity, the user recommendation score based on the likelihood score and the set of objective values;
providing, from the network service to the client device using the application program interface, user recommendation score for the entity; and
displaying, on a display of the client device using a processor on the client device, information about the entity in a format determined using the recommendation score of the entity.