Methods and apparatus for adjusting model threshold levels
Methods and apparatus for adjusting model threshold levels are disclosed. An example method includes determining volume estimates corresponding to a total number of users that satisfy a model threshold; when a ratio of (A) modeled users that satisfy the model threshold to (B) a sum of the modeled users that satisfy the model threshold and deterministic client device users does not satisfy a minimum ratio threshold, increasing the number of modeled users; when the total users that satisfy the model threshold does not satisfy a total volume target, increasing the number of modeled users; and updating the total volume target based on the increased number of modeled users; generating an adjusted count of users; and a model threshold adjuster to adjust the model threshold based on the adjusted count of users.
1. An apparatus comprising:
memory; and
one or more logic circuits to:
determine volume estimates corresponding to a total number of users that satisfy a model threshold within a first period of time;
when a ratio of (A) modeled users that satisfy the model threshold to (B) a sum of (i) the modeled users that satisfy the model threshold and (ii) deterministic client device users does not satisfy a minimum ratio threshold, increase the number of modeled users to satisfy the model threshold;
when the total users that satisfy the model threshold does not satisfy a total volume target, increase the number of modeled users to satisfy the total volume target; and
update the total volume target based on the increased number of modeled users;
apply an adjustment factor based on the volume estimates and the updated total volume target to an absolute count of users to generate an adjusted count of users; and
adjust the model threshold based on the adjusted count of users.
2. The apparatus of claim 1 , wherein the volume estimates include client device users and modeled client device users.
3. The apparatus of claim 1 , wherein the one or more logic circuits are to determine the volume estimates by determining a first volume estimate corresponding to a second total number of users that satisfy the model threshold within a second period of time and projecting the second total number of users based on a difference between the first period of time and the second period of time.
4. The apparatus of claim 1 , wherein the total volume target is a total number of users that satisfy the model threshold to be included in a data segment, the total number of users including client device users and modeled users.
5. The apparatus of claim 1 , wherein the updated total volume target is the sum of the increased number of modeled users that satisfy the model threshold and client device users that satisfy the model threshold.
6. The apparatus of claim 1 , wherein the one or more logic circuits are to generate the adjustment factor by dividing the updated total volume target by the volume estimates.
7. The apparatus of claim 1 , wherein the absolute count of users is a number of client device users and a number of model users that satisfy the model threshold within a second period of time.
8. The apparatus of claim 1 , wherein the one or more logic circuits are to adjust the model threshold based on the adjusted count of users by determining a model threshold value that, when applied to a model, changes the absolute count of users to a count nearest to the updated total volume target.
9. The apparatus of claim 1 , further including an interface to deploy a model corresponding the adjusted model threshold, the model corresponding to a data segment including adjusted count of users.
10. The apparatus of claim 9 , wherein the model further corresponds to a type of media, the deterministic client device users having been exposed to the type of media.
11. The apparatus of claim 1 , wherein the one or more logic circuits are to, when the ratio satisfies the minimum ratio threshold, maintain the total volume target.
12. A method comprising:
determining volume estimates corresponding to a total number of users that satisfy a model threshold within a first period of time;
when a ratio of (A) modeled users that satisfy the model threshold to (B) a sum of (i) the modeled users that satisfy the model threshold and (ii) deterministic client device users does not satisfy a minimum ratio threshold, increasing the number of modeled users to satisfy the model threshold;
when the total users that satisfy the model threshold does not satisfy a total volume target, increasing the number of modeled users to satisfy the total volume target;
updating the total volume target based on the increased number of modeled users;
applying an adjustment factor based on the volume estimates and the updated total volume target to an absolute count of users to generate an adjusted count of users; and
adjusting the model threshold based on the adjusted count of users.
13. The method of claim 12 , wherein the volume estimates include client device users and modeled client device users.
14. The method of claim 12 , wherein determining the volume estimates includes determining a first volume estimate corresponding to a second total number of users that satisfy the model threshold within a second period of time and projecting the second total number of users based on a difference between the first period of time and the second period of time.
15. The method of claim 12 , wherein the total volume target is a total number of users that satisfy the model threshold to be included in a data segment, the total number of users including client device users and modeled users.
16. The method of claim 12 , wherein the updated total volume target is the sum of the increased number of modeled users that satisfy the model threshold and client device users that satisfy the model threshold.
17. The method of claim 12 , further including generating the adjustment factor by dividing the updated total volume target by the volume estimates.
18. The method of claim 12 , wherein the absolute count of users is a number of client device users and a number of model users that satisfy the model threshold within a second period of time.
19. The method of claim 12 , wherein adjusting the model threshold based on the adjusted count of users includes determining a model threshold value that, when applied to a model, changes the absolute count of users to a count nearest to the updated total volume target.
20. A non-transitory computer readable medium comprising instructions which, when executed cause a machine to at least:
determine volume estimates corresponding to a total number of users that satisfy a model threshold within a first period of time;
when a ratio of (A) modeled users that satisfy the model threshold to (B) a sum of (i) the modeled users that satisfy the model threshold and (ii) deterministic client device users does not satisfy a minimum ratio threshold, increase the number of modeled users to satisfy the model threshold;
when the total users that satisfy the model threshold does not satisfy a total volume target, increase the number of modeled users to satisfy the total volume target;
update the total volume target based on the increased number of modeled users;
apply an adjustment factor based on the volume estimates and the updated total volume target to an absolute count of users to generate an adjusted count of users; and
adjust the model threshold based on the adjusted count of users.