Assigning synthetic respondents to geographic locations for audience measurement
View Patent ↗Example methods, apparatus, systems and articles of manufacture (e.g., physical storage media) to assign synthetic respondents to geographic locations for audience measurement are disclosed. Example apparatus disclosed herein include a constraint identifier to determine a set of constraints based on aggregate values of demographic features associated with the geographic locations. Disclosed example apparatus also include a parameter evaluator to determine a set of parameters of a model that is to return probabilities that respective ones of the respondents are associated with respective ones of the geographic locations, respective ones of the parameters being associated with respective ones of the constraints. Disclosed example apparatus further include a location estimator to evaluate the model based on the set of parameters and values of the demographic features for a first respondent to determine a set of probabilities that the first respondent is to be assigned to respective ones of the geographic locations.
1. An apparatus to assign virtual panelists to geographic locations for audience measurement, the apparatus comprising:
memory including computer readable instructions; and
a processor to execute the instructions to at least:
determine values of a set of parameters corresponding respectively to a set of demographic features associated with the geographic locations, the processor to determine the values of the set of parameters based on a first combination of aggregate values of the set of demographic features associated with the geographic locations and a second combination of respective sets of values of the set of demographic features for corresponding ones of the virtual panelists;
determine respective sets of assignment probabilities for the corresponding ones of the virtual panelists based on the determined values of the set of parameters and the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists; and
assign the virtual panelists to the geographic locations based on the sets of assignment probabilities.
2. The apparatus of claim 1 , wherein the set of demographic features associated with the geographic locations is a first set of demographic features, and the virtual panelists are associated with a second set of demographic features, the first set of demographic features missing at least one feature from the second set of demographic features.
3. The apparatus of claim 1 , wherein a first one of the geographic locations is associated with a first subset of the set of demographic features and a second one of the geographic locations is associated with a different second subset of the set of demographic features.
4. The apparatus of claim 3 , wherein a third one of the geographic locations is not associated with any of the features in the set of demographic features.
5. The apparatus of claim 1 , wherein the processor is to:
determine the first combination of aggregate values of the set of demographic features associated with the geographic locations based on a linear combination of the aggregate values of the set of demographic features multiplied with corresponding ones of the set of parameters; and
determine the second combination of the respective sets of values of the set of demographic features for corresponding ones of the virtual panelists based on (i) a first linear combination of a first set of values of the set of demographic features values for a first one of the virtual panelists multiplied by corresponding ones of the set of parameters and (ii) a second linear combination of a second set of values of the set of demographic features values for a second one of the virtual panelists multiplied by the corresponding ones of the set of parameters.
6. The apparatus of claim 1 , wherein the processor is to iteratively determine the set of parameters to minimize a difference between the first combination of the aggregate values of the set of demographic features associated with the geographic locations and the second combination of the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists.
7. The apparatus of claim 1 , wherein respective ones of the assignment probabilities in a first one of the sets of assignment probabilities for a first one of the virtual panelists represent how likely the first one of the virtual panelists is to belong in corresponding ones of the geographic locations.
8. A non-transitory computer readable medium comprising computer readable instructions which, when executed, cause a processor to at least:
determine values of a set of parameters corresponding respectively to a set of demographic features associated with a plurality of geographic locations, the values of the set of parameters to be determined based on a first combination of aggregate values of the set of demographic features associated with the geographic locations and a second combination of respective sets of values of the set of demographic features for corresponding ones of a plurality of virtual panelists;
determine respective sets of assignment probabilities for the corresponding ones of the virtual panelists based on the determined values of the set of parameters and the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists; and
assign the virtual panelists to the geographic locations based on the sets of assignment probabilities.
9. The non-transitory computer readable medium of claim 8 , wherein the set of demographic features associated with the geographic locations is a first set of demographic features, and the virtual panelists are associated with a second set of demographic features, the first set of demographic features missing at least one feature from the second set of demographic features.
10. The non-transitory computer readable medium of claim 8 , wherein a first one of the geographic locations is associated with a first subset of the set of demographic features and a second one of the geographic locations is associated with a different second subset of the set of demographic features.
11. The non-transitory computer readable medium of claim 10 , wherein a third one of the geographic locations is not associated with any of the features in the set of demographic features.
12. The non-transitory computer readable medium of claim 8 , wherein the instructions, when executed, cause the processor to:
determine the first combination of aggregate values of the set of demographic features associated with the geographic locations based on a linear combination of the aggregate values of the set of demographic features multiplied with corresponding ones of the set of parameters; and
determine the second combination of the respective sets of values of the set of demographic features for corresponding ones of the virtual panelists based on (i) a first linear combination of a first set of values of the set of demographic features values for a first one of the virtual panelists multiplied by corresponding ones of the set of parameters and (ii) a second linear combination of a second set of values of the set of demographic features values for a second one of the virtual panelists multiplied by the corresponding ones of the set of parameters.
13. The non-transitory computer readable medium of claim 8 , wherein the instructions, when executed, cause the processor to iteratively determine the set of parameters to minimize a difference between the first combination of the aggregate values of the set of demographic features associated with the geographic locations and the second combination of the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists.
14. The non-transitory computer readable medium of claim 8 , wherein respective ones of the assignment probabilities in a first one of the sets of assignment probabilities for a first one of the virtual panelists represent how likely the first one of the virtual panelists is to belong in corresponding ones of the geographic locations.
15. A method to assign virtual panelists to geographic locations for audience measurement, the method comprising:
determining, by executing an instruction with a processor, values of a set of parameters corresponding respectively to a set of demographic features associated with the geographic locations, the values of the set of parameters being determined based on a first combination of aggregate values of the set of demographic features associated with the geographic locations and a second combination of respective sets of values of the set of demographic features for corresponding ones of the virtual panelists;
determining, by executing an instruction with the processor, respective sets of assignment probabilities for the corresponding ones of the virtual panelists based on the determined values of the set of parameters and the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists; and
assigning, by executing an instruction with the processor, the virtual panelists to the geographic locations based on the sets of assignment probabilities.
16. The method of claim 15 , wherein the set of demographic features associated with the geographic locations is a first set of demographic features, and the virtual panelists are associated with a second set of demographic features, the first set of demographic features missing at least one feature from the second set of demographic features.
17. The method of claim 15 , wherein a first one of the geographic locations is associated with a first subset of the set of demographic features, a second one of the geographic locations is associated with a different second subset of the set of demographic features, and a third one of the geographic locations is not associated with any of the features in the set of demographic features.
18. The method of claim 16 , wherein the determining of the values of the set of parameters includes:
determining the first combination of aggregate values of the set of demographic features associated with the geographic locations based on a linear combination of the aggregate values of the set of demographic features multiplied with corresponding ones of the set of parameters; and
determining the second combination of the respective sets of values of the set of demographic features for corresponding ones of the virtual panelists based on (i) a first linear combination of a first set of values of the set of demographic features values for a first one of the virtual panelists multiplied by corresponding ones of the set of parameters and (ii) a second linear combination of a second set of values of the set of demographic features values for a second one of the virtual panelists multiplied by the corresponding ones of the set of parameters.
19. The method of claim 15 , wherein the determining of the values of the set of parameters includes iteratively determining the set of parameters to minimize a difference between the first combination of the aggregate values of the set of demographic features associated with the geographic locations and the second combination of the respective sets of values of the set of demographic features for the corresponding ones of the virtual panelists.
20. The method of claim 15 , wherein respective ones of the assignment probabilities in a first one of the sets of assignment probabilities for a first one of the virtual panelists represent how likely the first one of the virtual panelists is to belong in corresponding ones of the geographic locations.