US 11,756,070 B1
Predicting advertisement impact for campaign selection
Tianxiao Huang, San Francisco, CA (US)
Assigned to Quantcast Corporation, San Francisco, CA (US)
Filed by Quantcast Corporation, San Francisco, CA (US)
Filed on Sep. 14, 2020, as Appl. No. 17/19,967.
Application 17/019,967 is a continuation of application No. 16/002,436, filed on Jun. 7, 2018, granted, now 10,776,815.
Application 16/002,436 is a continuation of application No. 14/562,988, filed on Dec. 8, 2014, granted, now 10,019,728, issued on Jul. 10, 2018.
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0242 (2023.01); G06Q 30/0273 (2023.01)
CPC G06Q 30/0243 (2013.01) [G06Q 30/0273 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A method comprising:
constructing, by an influence system, an influence model for a campaign, comprising:
selecting a treatment group comprising a plurality of entities which been exposed to a advertising treatment of the campaign and;
selecting a control group comprising a plurality of entities, wherein the control group excludes entities of the treatment group and wherein the control group comprises entities which have not been exposed to the advertising treatment of the campaign;
selecting treatment group converters from the treatment group, each treatment group converter having met a conversion criterion the campaign after exposure to the campaign's advertising treatment;
selecting control group converters from the control group, each control group converter having met the conversion criterion of the campaign; and
comparing features of media consumption histories of the treatment group converters to features of media consumption histories of the control group converters;
receiving, by the influence system and from an auction system, a notification that an opportunity to expose a specified entity to advertising content is available, wherein the specified entity is not included in the treatment group;
responsive to receiving the notification, determining, by the influence system, a campaign frequency cap for the specified entity by applying the constructed influence model to features of the specified entity's media consumption history; and
providing, from the influence system to the auction system, a response configured according to the determined campaign frequency cap.