Inferring the presence of an entity within an identifier space based on the behaviors of unrelated entities
Embodiments of the invention build models to predict the likelihood of entities that operate in a given identifier space also operating in a disjoined identifier space based on a source panel of entities that operate in one or both of the identifier spaces. In operation, a model building engine builds a model based on features associated with the source panel and features associated with standard populations in the given identifier space. The model is used to determine whether the target entity is more similar to those entities in the source panel that operate only in the given identifier space or those entities in the source panel that operate in both identifier spaces.
1. A method comprising:
determining a first set of entities in a source panel operates in a first identifier space and not in a second identifier space;
determining a second set of entities in the source panel operates in the first identifier space and in the second identifier space;
building a model associated with the first identifier space based on a first set of features that distinguishes the first set of entities from a standard population associated with the first identifier space and a second set of features that distinguishes the second set of entities from the standard population; and
applying the model to a target entity that operates in the first identifier space to determine a likelihood of the target entity also operating in the second identifier space.
2. The method of claim 1 , wherein building the model associated with the first identifier space comprises:
identifying the first set of features by comparing a set of features associated with the first set of entities with a set of features associated with the standard population; and
identifying the second set of features by comparing a set of features associated with the second set of entities with the set of features associated with the standard population.
3. The method of claim 2 , wherein the standard population is a baseline population within the first identifier space, and the set of features associated with the standard population comprises features of the baseline population determined based on operations of the baseline population within the identifier space.
4. The method of claim 2 , wherein the standard population is a general population, and the set of features associated with the standard population comprises features determined based on census data.
5. The method of claim 1 , wherein applying the model to the target entity comprises:
computing a first score indicative of the similarity between the target entity and the first set of entities by comparing a consumption history associated with the target entity with the first set of features;
computing a second score indicative of the similarity between the target entity and the second set of entities by comparing a consumption history associated with the target entity with the second set of features;
comparing the first score and the second score to determine the likelihood of the target entity also operating in the second identifier space.
6. The method of claim 5 , wherein comparing the first score and the second score comprises determining which of the first score and the second score is higher, and the first score being higher than the second score indicates that the likelihood of the target entity also operating in the second identifier space is lower than when the second score is higher than the first score.
7. The method of claim 1 , wherein each entity operating in the first identifier space is associated with a unique identifier specific to the first identifier space, and each entity operating in the second identifier space is associated with a unique identifier specific to the second identifier space.
8. The method of claim 7 , wherein for each entity in the second set of entities, a unique identifier associated with the entity and specific to the first identifier space is mapped to a unique identifier associated with the entity and specific to the second identifier space.
9. The method of claim 7 , wherein the target entity operating in the first identifier space also operates in the second identifier space, and a unique identifier associated with the target entity and specific to the first identifier space is not mapped to a unique identifier associated with the target entity and specific to the second identifier space.
10. The method of claim 1 , wherein determining that the first set of entities in the source panel operates in a first identifier space and not in the second identifier space comprises determining based on an inference that the first set of entities operates in the first identifier space and not in the second identifier space.
11. The method of claim 1 , wherein determining that the second set of entities in the source panel operates in the first identifier space and in the second identifier space comprises determining based on an inference that the second set of entities operates in the first identifier space and in the second identifier space.
12. A non-transitory computer readable storage medium storing processor-executable instructions, the instructions configured to be executed by the processor to implement a method comprising:
determining a first set of entities in a source panel operates in a first identifier space and not in a second identifier space;
determining a second set of entities in the source panel operates in the first identifier space and in the second identifier space;
building a model associated with the first identifier space based on a first set of features that distinguishes the first set of entities from a standard population associated with the first identifier space and a second set of features that distinguishes the second set of entities from the standard population; and
applying the model to a target entity that operates in the first identifier space to determine a likelihood of the target entity also operating in the second identifier space.
13. The medium of claim 12 , wherein building the model associated with the first identifier space comprises:
identifying the first set of features by comparing a set of features associated with the first set of entities with a set of features associated with the standard population; and
identifying the second set of features by comparing a set of features associated with the second set of entities with the set of features associated with the standard population.
14. The medium of claim 13 , wherein the standard population is a baseline population within the first identifier space, and the set of features associated with the standard population comprises features of the baseline population determined based on operations of the baseline population within the identifier space.
15. The medium of claim 13 , wherein the standard population is a general population, and the set of features associated with the standard population comprises features determined based on census data.
16. The medium of claim 12 , wherein applying the model to the target entity comprises:
computing a first score indicative of the similarity between the target entity and the first set of entities by comparing a consumption history associated with the target entity with the first set of features;
computing a second score indicative of the similarity between the target entity and the second set of entities by comparing a consumption history associated with the target entity with the second set of features;
comparing the first score and the second score to determine the likelihood of the target entity also operating in the second identifier space.
17. The medium of claim 16 , wherein comparing the first score and the second score comprises determining which of the first score and the second score is higher, and the first score being higher than the second score indicates that the likelihood of the target entity also operating in the second identifier space is lower than when the second score is higher than the first score.
18. The medium of claim 12 , wherein each entity operating in the first identifier space is associated with a unique identifier specific to the first identifier space, and each entity operating in the second identifier space is associated with a unique identifier specific to the second identifier space.
19. The medium of claim 18 , wherein for each entity in the second set of entities, a unique identifier associated with the entity and specific to the first identifier space is mapped to a unique identifier associated with the entity and specific to the second identifier space.
20. The medium of claim 18 , wherein the target entity operating in the first identifier space also operates in the second identifier space, and a unique identifier associated with the target entity and specific to the first identifier space is not mapped to a unique identifier associated with the target entity and specific to the second identifier space.
21. The medium of claim 12 , wherein determining that the first set of entities in the source panel operates in a first identifier space and not in the second identifier space comprises determining based on an inference that the first set of entities operates in the first identifier space and not in the second identifier space.
22. The medium of claim 12 , wherein determining that the second set of entities in the source panel operates in the first identifier space and in the second identifier space comprises determining based on an inference that the second set of entities operates in the first identifier space and in the second identifier space.
23. A system comprising:
a processor; and
a computer readable storage medium storing processor-executable computer program instructions, the computer program instructions comprising instructions for:
determining a first set of entities in a source panel operates in a first identifier space and not in a second identifier space;
determining a second set of entities in the source panel operates in the first identifier space and in the second identifier space;
building a model associated with the first identifier space based on a first set of features that distinguishes the first set of entities from a standard population associated with the first identifier space and a second set of features that distinguishes the second set of entities from the standard population; and
applying the model to a target entity that operates in the first identifier space to determine a likelihood of the target entity also operating in the second identifier space.