Back to results
US 9,922,333 B2Granted
Automated multivariate behavioral prediction
AssigneeExponential Interactive, Inc.
InventorAlexander Saldanha
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.
Claims32
A computer-implemented method for developing an advertiser-specific behavioral lift model, the method executed by a computer system having a processor and a data store and comprising:
- maintaining, by the processor, in the data store a plurality of behavior profiles, each behavior profile including a persistent identifier and a plurality of behavior vectors for the persistent identifier, the behavior vectors indicating occurrences of each of a set of behaviors associated with the persistent identifier, each occurrence having a date;
- receiving, by the processor, conversion information relating to an advertiser associated with the plurality of behavior profiles, the conversion information indicating responses to an advertisement of the advertiser, the responses associated with user profiles;
- calculating by the processor a sequence of behavior tuples based upon the behavioral vectors and the conversion information, the calculated sequence including a first position and a last position, the calculation of the sequence of behavior tuples including:
- identifying, by the processor, a population of behavior profiles for each position in the sequence proceeding from the first position to the last position, the population associated with every position excluding behavior profiles that include behavior vectors indicating occurrences of any behavior tuple associated with any previous position in the sequence between the given position and the first behavior tuple in the sequence;
- and
- selecting, by the processor, a behavior tuple for each position in the sequence based on a lift for the behavior tuple relative to the population of behavior profiles identified for the position, the lift calculated relative to the conversion information;
- and
- storing in the data store, by the processor, the sequence of behavior tuples as an advertiser-specific behavioral lift model.
5 more elements
A computer-implemented method for developing an advertiser-specific behavioral lift model, the method executed by a computer system having a processor and a data store and comprising:
- maintaining, by the processor, in the data store a plurality of behavior profiles, each behavior profile including a persistent identifier of a user and indicating occurrences of each of a set of behaviors associated with the persistent identifier;
- receiving, by the processor, conversion information relating to an advertiser associated with the plurality of behavior profiles, the conversion information indicating responses to an advertisement of the advertiser, the responses associated with behavior profiles;
- calculating by the processor a sequence of behavior tuples based upon the behavior profiles and the conversion information, the calculated sequence including a first position and a last position, the calculation of the sequence of behavior tuples including;
- identifying, by the processor, a population of behavior profiles for each position in the sequence proceeding from the first position to the last position, the population associated with every position excluding behavior profiles that include behaviors in the set of behaviors indicating occurrences of any behavior tuple associated with any previous position in the sequence between the given position and the first behavior tuple in the sequence;
- and
- selecting, by the processor, a behavior tuple for each position in the sequence based on a lift for the behavior tuple relative to the population of behavior profiles identified for the position, the lift calculated relative to the conversion information;
- and
- storing in the data store by the processor the sequence of behavior tuples as an advertiser-specific behavioral lift model.
5 more elements
A non-transitory computer-readable medium for developing an advertiser-specific behavioral lift model comprising instructions for execution on a processor, which when executed cause the processor to perform steps of:
- maintaining in a data store a plurality of user profiles, each user profile including a persistent identifier of a user and indicating occurrences of each of a set of behaviors by the user;
- receiving conversion information relating to an advertiser associated with the plurality of user profiles, the conversion information indicating users' responses to an advertisement of the advertiser, the users' responses associated with user profiles;
- calculating a sequence of behavior tuples based upon the user profiles and the conversion information, the calculated sequence including a first position and a last position, the calculation of the sequence of behavior tuples including;
- identifying, by the processor, a population of user profiles for each position in the sequence proceeding from the first position to the last position, the population associated with every position excluding user profiles that include behaviors in the set of behaviors indicating occurrences of any behavior tuple associated with any previous position in the sequence between the given position and the first behavior tuple in the sequence;
- and
- selecting a behavior tuple for each position in the sequence based on a lift for the behavior tuple relative to the population of user profiles identified for the position, the lift calculated relative to the conversion information;
- and
- storing in the data store the sequence of behavior tuples as an advertiser-specific behavioral lift model.
5 more elements
A computing system for developing an advertiser-specific behavioral lift model, comprising:
- a processor capable of executing a set of instructions;
- a data storage containing a plurality of user profiles, each user profile including a persistent identifier of a user and indicating occurrences of each of a set of behaviors by a user;
- a memory storing instructions for execution by the processor, the instructions causing the processor to perform steps of:
- receiving conversion information relating to an advertiser associated with the plurality of user profiles, the conversion information indicating users' responses to an advertisement of the advertiser, the users' responses associated with user profiles;
- calculating a sequence of behavior tuples based upon the user profiles and the conversion information, the calculated sequence including a first position and a last position, the calculation of the sequence of behavior tuples including;
- identifying in the data storage a population of user profiles for each position in the sequence proceeding from the first position to the last position, the population associated with every position excluding user profiles that include behaviors in the set of behaviors indicating occurrences of any behavior tuple associated with any previous position in the sequence between the given position and the first behavior tuple in the sequence;
- and
- selecting a behavior tuple for each position in the sequence based on a lift for the behavior tuple relative to the population of user profiles identified for the position, the lift calculated relative to the conversion information;
- and
- storing in the data storage the sequence of behavior tuples as an advertiser-specific behavioral lift model.
7 more elements
Dependencies are inferred from the claim text.