Methods and apparatus to correct misattributions of media impressions
Methods, apparatus, and articles of manufacture to correct misattributions of media impressions are disclosed. An example method includes obtaining first demographic-based impressions via a beacon transmitted in response to access to content by a first set of panelists, obtaining, from a database proprietor, second demographic-based impressions of the content on a second set of persons, forming a pseudo-inverse matrix determined based in part on the first impressions, and having a truncated value and a damped value to form third demographic-based impressions of the content on the second set of persons based on the second impressions, and computing at least partially corrected demographic-based impressions values by multiplying a vector of database proprietor impression data by the pseudo-inverse matrix.
1. A method of correcting demographic-based impressions of content provided by a content provider site over a network, the method comprising:
obtaining first demographic-based impressions via a beacon transmitted via a network in response to access to content by a first set of panelists;
obtaining, from a database proprietor, second demographic-based impressions of the content on a second set of persons;
forming a pseudo-inverse matrix determined based in part on the first demographic-based impressions, and having a truncated value and a damped value to form third demographic-based impressions of the content on the second set of persons based on the second demographic-based impressions;
computing at least partially corrected demographic-based impression values by multiplying a vector of database proprietor impression data by the pseudo-inverse matrix to determine misattribution correction factors improving accuracy of impression-based data including at least one of a unique audience or impression count to permit a computer or processing system to produce audience analysis information with higher accuracy.
2. The method as defined in claim 1 , wherein the database proprietor includes a service provider and a social network service.
3. The method as defined in claim 1 , wherein the truncated value is calculated by applying a small threshold to the pseudo-inverse matrix.
4. The method as defined in claim 1 , wherein the damped value is calculated by applying a damping factor to the pseudo-inverse matrix.
5. The method as defined in claim 4 , wherein the damping factor is a percentage of a maximum singular value of the pseudo-inverse matrix.
6. The method as defined in claim 1 , wherein a replacement impression value is imputed for at least one of the third demographic-based impressions.
7. The method as defined in claim 1 , wherein the pseudo-inverse matrix is determined using Singular Value Decomposition (SVD).
8. An apparatus to correct demographic-based impressions of content provided by a content provider site over a network, the apparatus comprising a misattribution corrector to:
obtain first demographic-based impressions via a beacon transmitted via a network in response to access to content by a first set of panelists;
obtain, from a database proprietor, second demographic-based impressions of the content on a second set of persons;
form a pseudo-inverse matrix determined based in part on the first demographic-based impressions, and having a truncated value and a damped value to form third demographic-based impressions of the content on the second set of persons based on the second demographic-based impressions;
compute at least partially corrected demographic-based impression values by multiplying a vector of database proprietor impression data by the pseudo-inverse matrix; and
determine misattribution correction factors to improve accuracy of impression-based data including at least one of a unique audience or impression count to permit a computer or processing system to produce audience analysis information with higher accuracy.
9. The apparatus of claim 8 , wherein the database proprietor includes a service provider and a social network service.
10. The apparatus of claim 8 , wherein the truncated value is calculated by applying a small threshold to the pseudo-inverse matrix.
11. The apparatus of claim 8 , wherein the damped value is calculated by applying a damping factor to the pseudo-inverse matrix.
12. The apparatus of claim 11 , wherein the damping factor is a percentage of a maximum singular value of the pseudo-inverse matrix.
13. The apparatus of claim 8 , wherein a replacement impression value is imputed for at least one of the third demographic-based impressions.
14. The apparatus of claim 8 , wherein the pseudo-inverse matrix is determined using Singular Value Decomposition (SVD).
15. A non-transitory computer-readable storage medium comprising instructions that, when executed, cause a machine to at least:
obtain first demographic-based impressions via a beacon transmitted via a network in response to access to content by a first set of panelists;
obtain, from a database proprietor, second demographic-based impressions of the content on a second set of persons;
form a pseudo-inverse matrix determined based in part on the first demographic-based impressions, and having a truncated value and a damped value to form third demographic-based impressions of the content on the second set of persons based on the second demographic-based impressions;
compute at least partially corrected demographic-based impressions values by multiplying a vector of database proprietor impression data by the pseudo-inverse matrix; and
determine misattribution correction factors to improve accuracy of impression-based data including at least one of a unique audience or impression count to permit a computer or processing system to produce audience analysis information with higher accuracy.
16. The non-transitory computer-readable storage medium as defined in claim 15 , wherein the truncated value is calculated by applying a small threshold to the pseudo-inverse matrix.
17. The non-transitory computer-readable storage medium as defined in claim 15 , wherein the damped value is calculated by applying a damping factor to the pseudo-inverse matrix.
18. The non-transitory computer-readable storage medium as defined in claim 17 , wherein the damping factor is a percentage of a maximum singular value of the pseudo-inverse matrix.
19. The non-transitory computer-readable storage medium as defined in claim 15 , wherein a replacement impression value is imputed for at least one of the third demographic-based impressions.
20. The non-transitory computer-readable storage medium as defined in claim 15 , wherein the pseudo-inverse matrix is determined using Singular Value Decomposition (SVD).