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 media provided by a provider site over a network, the method comprising:
obtaining first demographic-based impressions transmitted via a network in response to access to media by a first set of panelists;
obtaining, from a database proprietor, second demographic-based impressions of the media accessed by a second set of persons;
forming a pseudo-inverse matrix determined based in part on the first demographic-based impressions;
forming third demographic-based impressions of the media based on the second demographic-based impressions; and
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.
2. The method of claim 1 , wherein forming third demographic-based impressions includes using a truncated value of the pseudo-inverse matrix, the truncated value based on a threshold applied to the pseudo-inverse matrix.
3. The method of claim 1 , wherein forming third demographic-based impressions includes applying a damping factor to the pseudo-inverse matrix.
4. The method of claim 3 , wherein the damping factor is a percentage of a maximum singular value of the pseudo-inverse matrix, the percentage of the maximum singular value any percentage resulting in tuning of remaining singular values of the pseudo-inverse matrix.
5. The method of claim 4 , further including damping a singular value of the pseudo-inverse matrix to reduce a high noise component resulting from truncation of the pseudo-inverse matrix.
6. The method of claim 5 , wherein the singular value is damped when the singular value is determined to be greater than the damping factor.
7. The method of claim 1 , wherein forming the pseudo-inverse matrix includes replacing a non-zero diagonal entry with its reciprocal and transposing the resulting matrix.
8. An apparatus to correct demographic-based impressions of media provided by a provider site over a network, the apparatus comprising a misattribution corrector to:
obtain first demographic-based impressions transmitted via a network in response to access to media by a first set of panelists;
obtain, from a database proprietor, second demographic-based impressions of the media accessed by a second set of persons;
form a pseudo-inverse matrix determined based in part on the first demographic-based impressions;
form third demographic-based impressions of the media based on the second demographic-based impressions; and
compute 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.
9. The apparatus of claim 8 , wherein a truncated value of the pseudo-inverse matrix is used to form the third demographic-based impressions, the truncated value based on a threshold applied to the pseudo-inverse matrix.
10. The apparatus of claim 8 , wherein a damping factor is applied to the pseudo-inverse matrix.
11. The apparatus of claim 10 , wherein the damping factor is a percentage of a maximum singular value of the pseudo-inverse matrix, the percentage of the maximum singular value any percentage resulting in tuning of remaining singular values of the pseudo-inverse matrix.
12. The apparatus of claim 11 , wherein a singular value of the pseudo-inverse matrix is damped to reduce a high noise component resulting from truncation of the pseudo-inverse matrix.
13. The apparatus of claim 12 , wherein the singular value is damped when the singular value is determined to be greater than the damping factor.
14. The apparatus of claim 8 , wherein the pseudo-inverse matrix is formed by replacing a non-zero diagonal entry with its reciprocal and transposing the resulting matrix.
15. A non-transitory computer-readable storage medium comprising instructions that, when executed, cause a machine to at least:
obtain first demographic-based impressions transmitted via a network in response to access to media by a first set of panelists;
obtain, from a database proprietor, second demographic-based impressions of the media accessed by a second set of persons;
form a pseudo-inverse matrix determined based in part on the first demographic-based impressions;
form third demographic-based impressions of the media based on the second demographic-based impressions; and
compute 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.
16. The non-transitory computer-readable storage medium as defined in claim 15 , wherein a truncated value of the pseudo-inverse matrix is used to form the third demographic-based impressions, the truncated value based on a threshold applied to the pseudo-inverse matrix.
17. The non-transitory computer-readable storage medium as defined in claim 15 , wherein a damping factor is applied 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, the percentage of the maximum singular value any percentage resulting in tuning of remaining singular values of the pseudo-inverse matrix.
19. The non-transitory computer-readable storage medium as defined in claim 18 , wherein a singular value of the pseudo-inverse matrix is damped to reduce a high noise component resulting from truncation of the pseudo-inverse matrix.
20. The non-transitory computer-readable storage medium as defined in claim 19 , wherein the singular value is damped when the singular value is determined to be greater than the damping factor.
21. The non-transitory computer-readable storage medium as defined in claim 15 , wherein the pseudo-inverse matrix is formed by replacing a non-zero diagonal entry with its reciprocal and transposing the resulting matrix.