Granted Patent B1

US 11,507,604 · App. 16/987,755 · Granted Nov 22, 2022

Inventor: William Kennedy Browne (San Francisco, CA)

Assignee: Quantcast Corporation

G06F16/285G06F16/958

Abstract

Claims (74)

1. A method, comprising:

receiving histories of a plurality of entities, each respective history comprising a plurality of features, each entity assigned to at least one cluster from a plurality of clusters;

for each of the plurality of clusters, generating a corresponding cluster feature list, each cluster feature list comprising features from the histories corresponding to entities assigned to the respective cluster;

for each of the features comprised in the plurality of cluster feature lists, determining a corresponding variance coefficient across the plurality of clusters, each variance coefficient corresponding to the variance of each of the features across the plurality of clusters;

for each of the plurality of clusters, generating a respective cluster variance feature list comprising a plurality of cluster variance features, the plurality of cluster variance features corresponding to features in the respective cluster feature list having a respective variance coefficient greater than a variance threshold value;

for each of the cluster variance features corresponding to each cluster variance feature list, determining a corresponding general prevalence coefficient, each general prevalence coefficient representing a prevalence of entities in the plurality of entities having the respective cluster variance feature in their history;

for each of the plurality of clusters, determining an in-cluster prevalence coefficient corresponding to each cluster variance feature in the respective cluster variance feature list, the in-cluster prevalence coefficient representing a prevalence of entities in the respective cluster having the respective cluster variance feature in their history;

for each cluster, generating a respective discriminating feature list comprising a plurality of discriminating features, by selecting cluster variance features corresponding to the respective cluster having a difference between each cluster variance feature's corresponding in-cluster prevalence coefficient and the corresponding general prevalence coefficient greater than a prevalence coefficient threshold;

choosing at least one cluster by comparing features of a history of a specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters;

customizing content according to the at least one chosen cluster; and

sending the customized content to the specified entity.

2. The method of claim 1 , further comprising:

assigning each received entity to at least one cluster from the plurality of clusters.

3. The method of claim 1 , wherein:

choosing at least one cluster further comprises identifying at least one persona corresponding to the specified entity by comparing features of the history of the specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters; and

customizing content according to the at least one chosen cluster further comprises customizing content according to the at least one identified persona.

4. The method of claim 1 , wherein customizing content according to the at least one chosen cluster comprises:

selecting an item of content associated with the at least one chosen cluster.

5. The method of claim 1 , wherein customizing content according to the at least one chosen cluster comprises:

customizing an item of content associated with the at least one chosen cluster.

6. The method of claim 1 , wherein customizing content according to the at least one chosen cluster comprises:

providing customization instructions associated with the at least one chosen cluster.

7. The method of claim 6 , further comprising:

providing customization instructions associated with the at least one chosen cluster to an ad server.

8. A non-transitory computer-readable storage medium storing processor-executable computer program instructions that, when executed, cause a computer processor to perform a method, the method comprising:

receiving histories of a plurality of entities, each respective history comprising a plurality of features, each entity assigned to at least one cluster from a plurality of clusters;

for each of the plurality of clusters, generating a corresponding cluster feature list, each cluster feature list comprising features from the histories corresponding to entities assigned to the respective cluster;

for each of the features comprised in the plurality of cluster feature lists, determining a corresponding variance coefficient across the plurality of clusters, each variance coefficient corresponding to the variance of each of the features across the plurality of clusters;

for each of the plurality of clusters, generating a respective cluster variance feature list comprising a plurality of cluster variance features, the plurality of cluster variance features corresponding to features in the respective cluster feature list having a respective variance coefficient greater than a variance threshold value;

for each of the cluster variance features corresponding to each cluster variance feature list, determining a corresponding general prevalence coefficient, each general prevalence coefficient representing a prevalence of entities in the plurality of entities having the respective cluster variance feature in their history;

for each of the plurality of clusters, determining an in-cluster prevalence coefficient corresponding to each cluster variance feature in the respective cluster variance feature list, the in-cluster prevalence coefficient representing a prevalence of entities in the respective cluster having the respective cluster variance feature in their history;

for each cluster, generating a respective discriminating feature list comprising a plurality of discriminating features, by selecting cluster variance features corresponding to the respective cluster having a difference between each cluster variance feature's corresponding in-cluster prevalence coefficient and the corresponding general prevalence coefficient greater than a prevalence coefficient threshold;

choosing at least one cluster by comparing features of a history of a specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters;

customizing content according to the at least one chosen cluster; and

sending the customized content to the specified entity.

9. The non-transitory computer-readable storage medium of claim 8 , wherein the method further comprises:

assigning each received entity to at least one cluster from the plurality of clusters.

10. The non-transitory computer-readable storage medium of claim 8 , wherein:

choosing at least one cluster further comprises identifying at least one persona corresponding to the specified entity by comparing features of the history of the specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters; and

customizing content according to the at least one chosen cluster further comprises customizing content according to the at least one identified persona.

11. The non-transitory computer-readable storage medium of claim 8 , wherein customizing content according to the at least one chosen cluster comprises:

selecting an item of content associated with the at least one chosen cluster.

12. The non-transitory computer-readable storage medium of claim 8 , wherein customizing content according to the at least one chosen cluster comprises:

customizing an item of content associated with the at least one chosen cluster.

13. The non-transitory computer-readable storage medium of claim 8 , wherein customizing content according to the at least one chosen cluster comprises:

providing customization instructions associated with the at least one chosen cluster.

14. The non-transitory computer-readable storage medium of claim 13 , further comprising:

providing customization instructions associated with the at least one chosen cluster to an ad server.

15. A system comprising:

a processor; and

a non-transitory computer-readable storage medium storing processor-executable computer program instructions that, when executed, cause a computer processor to perform a method, the method comprising:

receiving histories of a plurality of entities, each respective history comprising a plurality of features, each entity assigned to at least one cluster from a plurality of clusters;

for each of the plurality of clusters, generating a corresponding cluster feature list, each cluster feature list comprising features from the histories corresponding to entities assigned to the respective cluster;

for each of the features comprised in the plurality of cluster feature lists, determining a corresponding variance coefficient across the plurality of clusters, each variance coefficient corresponding to the variance of each of the features across the plurality of clusters;

for each of the plurality of clusters, generating a respective cluster variance feature list comprising a plurality of cluster variance features, the plurality of cluster variance features corresponding to features in the respective cluster feature list having a respective variance coefficient greater than a variance threshold value;

for each of the cluster variance features corresponding to each cluster variance feature list, determining a corresponding general prevalence coefficient, each general prevalence coefficient representing a prevalence of entities in the plurality of entities having the respective cluster variance feature in their history;

for each of the plurality of clusters, determining an in-cluster prevalence coefficient corresponding to each cluster variance feature in the respective cluster variance feature list, the in-cluster prevalence coefficient representing a prevalence of entities in the respective cluster having the respective cluster variance feature in their history;

for each cluster, generating a respective discriminating feature list comprising a plurality of discriminating features, by selecting cluster variance features corresponding to the respective cluster having a difference between each cluster variance feature's corresponding in-cluster prevalence coefficient and the corresponding general prevalence coefficient greater than a prevalence coefficient threshold;

choosing at least one cluster by comparing features of a history of a specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters;

customizing content according to the at least one chosen cluster; and

sending the customized content to the specified entity.

16. The system of claim 15 , wherein the method further comprises:

assigning each received entity to at least one cluster from the plurality of clusters.

17. The system of claim 15 , wherein:

choosing at least one cluster further comprises identifying at least one persona corresponding to the specified entity by comparing features of the history of the specified entity to the plurality of discriminating features of the respective discriminating feature list corresponding to each of the plurality of clusters; and

customizing content according to the at least one chosen cluster further comprises customizing content according to the at least one identified persona.

18. The system of claim 15 , wherein customizing content according to the at least one chosen cluster comprises:

selecting an item of content associated with the at least one chosen cluster.

19. The system of claim 15 , wherein customizing content according to the at least one chosen cluster comprises:

customizing an item of content associated with the at least one chosen cluster.

20. The system of claim 15 , wherein customizing content according to the at least one chosen cluster comprises:

providing customization instructions associated with the at least one chosen cluster.

21. The system of claim 20 , further comprising:

providing customization instructions associated with the at least one chosen cluster to an ad server.

Assignments (5)

Continuity (1)

Cited By (13)

US 12,212,079 US 12,244,078 US 12,255,412 US 12,272,884 US 12,278,433 US 12,316,027 US 12,519,233 US 12,542,355 US 12,548,904 US 12,555,908 US 12,609,449 US 12,627,050 US 12,627,051