Using proxy behaviors for audience selection
Method and system for assessing the suitability of an entity using a proxy. A description of a behavior associated with a desirable audience is received. A proxy behavior estimated to be characteristic of the desirable audience is selected. The proxy behavior comprises the performance of proxy events related to the consumption of media received by an entity over a network, which can be found in an entity's consumption history. An entity can be assessed for inclusion in a proxy audience, by examining the entity's consumption history for proxy behaviors. A behavioral model is built using a training set comprising the proxy audience. By applying the behavioral model to the consumption history of a specified entity, the specified entity's suitability for selection can be determined. Advantageously, in an embodiment, the invention enables the use of behavioral modeling techniques even when the complete behavior of the desirable audience is not available.
1. A computer-implemented method comprising:
receiving focus information descriptive of a search engine advertising campaign from an advertising campaign manager, the focus information comprising two or more keyword search strings, each keyword search string paired with at least one bid parameter, an importance of each keyword search string characterized by the at least one bid parameter;
accessing a respective consumption history of each of a plurality of entities in a storage;
identifying, at an audience selection system, a set of entities from the storage by examining each of the respective consumption histories for one or more proxy events, each proxy event comprising a keyword search matching at least one of the two or more keyword search strings described in the focus information;
weighting each proxy event according to the at least one bid parameter paired with each proxy event's keyword search string;
creating a training set by selecting entities from the set of entities for inclusion in a training set based on the proxy event weights of the one or more proxy events found in each entity's respective consumption history;
building a behavioral model based on the training set;
receiving a specified entity's consumption history; and
assessing the suitability of the specific entity for selection by applying the behavioral model to the specified entity's consumption history.
2. The method of claim 1 wherein:
a relative importance of each keyword search string is characterized by the at least one bid parameter.
3. The method of claim 1 wherein:
the bid parameter increases the weight of the proxy event.
4. The method of claim 1 wherein:
the bid parameter decreases the weight of the proxy event.
5. The method of claim 1 wherein:
the bid parameter comprises a maximum bid.
6. The method of claim 1 further comprising
weighting at least one proxy event according to a recency of the at least one proxy event.
7. The method of claim 1 further comprising
weighting at least one proxy event according to a frequency of the at least one proxy event.
8. A non-transitory computer readable storage medium including computer program instructions that when executed, cause a computer processor to perform operations comprising:
receiving focus information descriptive of a search engine advertising campaign from an advertising campaign manager, the focus information comprising two or more keyword search strings, each keyword search string paired with at least one bid parameter, an importance of each keyword search string characterized by the at least one bid parameter;
accessing a respective consumption history of each of a plurality of entities in a storage;
identifying, at an audience selection system, a set of entities from the storage by examining each of the respective consumption histories for one or more proxy events, each proxy event comprising a keyword search matching at least one of the two or more keyword search strings described in the focus information;
weighting each proxy event according to the at least one bid parameter paired with each proxy event's keyword search string;
creating a training set by selecting entities from the set of entities for inclusion in a training set based on the proxy event weights of the one or more proxy events found in each entity's respective consumption history;
building a behavioral model based on the training set;
receiving a specified entity's consumption history; and
assessing the suitability of the specific entity for selection by applying the behavioral model to the specified entity's consumption history.
9. The medium of claim 8 wherein:
the bid parameter increases the weight of the proxy event.
10. The medium of claim 8 wherein:
the bid parameter decreases the weight of the proxy event.
11. The medium of claim 8 wherein:
the bid parameter comprises a maximum bid.
12. The medium of claim 8 wherein the operation of weighting further comprises:
weighting at least one proxy event according to a recency of the at least one proxy event.
13. The medium of claim 8 wherein the operation of weighting further comprises:
weighting at least one proxy event according to a frequency of the at least one proxy event.
14. A system comprising:
a processor; and
a non-transitory computer readable storage medium storing processor-executable computer program instructions that when executed, cause a computer processor to perform operations comprising:
receiving focus information descriptive of a search engine advertising campaign from an advertising campaign manager, the focus information comprising two or more keyword search strings, each keyword search string paired with at least one bid parameter, an importance of each keyword search string characterized by the at least one bid parameter;
accessing a respective consumption history of each of a plurality of entities in a storage;
identifying, at an audience selection system, a set of entities from the storage by examining each of the respective consumption histories for one or more proxy events, each proxy event comprising a keyword search matching at least one of the two or more keyword search strings described in the focus information;
weighting each proxy event according to the at least one bid parameter paired with each proxy event's keyword search string;
creating a training set by selecting entities from the set of entities for inclusion in a training set based on the proxy event weights of the one or more proxy events found in each entity's respective consumption history;
building a behavioral model based on the training set;
receiving a specified entity's consumption history; and
assessing the suitability of the specific entity for selection by applying the behavioral model to the specified entity's consumption history.
15. The system of claim 14 wherein:
the bid parameter increases the weight of the proxy event.
16. The system of claim 14 wherein:
the bid parameter decreases the weight of the proxy event.
17. The system of claim 14 wherein:
the bid parameter comprises a maximum bid.
18. The system of claim 14 wherein the operation of weighting further comprises:
weighting at least one proxy event according to a recency of the at least one proxy event.
19. The system of claim 14 wherein the operation of weighting further comprises:
weighting at least one proxy event according to a frequency of the at least one proxy event.