Advertisement selection using multivariate behavioral model
AssigneeExponential Interactive, Inc.
InventorsAlexander Saldanha, Mikhail Faiguenblat
Life of this patent
Estimating the term…
Abstract
An advertising system identifies behaviors from user activity and associates the behaviors with a user profile. Advertisers provide the advertising system with information on conversion rates of users associated with user profiles. A behavioral model of user responses is built to identify the relative frequency of behaviors for increasing the response rate of ads. Incoming advertising requests are matched to modeled behaviors to determine an advertiser's interest in bidding on the ad placement.
Claims12
A computer-implemented method for identifying the value of bidding on an advertisement to be placed to a user, the method executed by a computer system, and comprising:
- maintaining, in a database by the computer system, a plurality of lift curves associated with a plurality of advertisers, wherein the lift curve for each advertiser identifies, relative to that advertiser's conversions, a sequence of behavior tuples, each behavior tuple in the sequence describing at least one behavior and associated with a lift reflecting a relative likelihood of conversion for the incremental population of persistent identifiers reached by each behavior tuple relative to a baseline excluding the population of persistent identifiers targeted by all prior behavior tuples in the sequence, the behavior tuples in the sequence ordered from greatest to least amount of associated lift;
- maintaining, in the database by the computer system, for each advertiser, advertising flight information including an estimated value to the advertiser for a conversion and an average conversion rate for the advertiser;
- receiving, by the computer system, a request for a placement of an advertisement from a user device, the request including a persistent identifier of the user device, and a context for the placement;
- identifying, by the computer system, a set of behaviors associated with the persistent identifier of the user device;
- selecting advertisers to bid on providing an advertisement to the user device in response to the context of the placement;
- determining, by the computer system, the lift for each of the selected advertisers relative to the lift curve associated with the selected advertiser, wherein the lift is determined by identifying, in the sequence of behavior tuples in the lift curve for the selected advertiser, a single behavior tuple associated with the highest lift that corresponds to the behaviors associated with the persistent identifier;
- determining, by the computer system, for each selected advertiser a placement value for the placement of an advertisement to the user device, using the advertiser's estimated value for a conversion, and the average conversion rate for the advertiser and only the lift associated with the identified single behavior tuple from the sequence of behavior tuples in that advertiser's lift curve;
- determining, by the computer system, a bid for each selected advertiser using the placement value for the selected advertiser;
- determining, by the computer system, an advertiser providing a highest bid;
- selecting, by the computer system, an advertisement from the advertiser providing a highest bid;
- obtaining from a database, the selected advertisement;
- and
- providing, by the computer system, the selected advertisement to the user device.
Dependencies are inferred from the claim text.