Methods and apparatus to accurately credit streaming sessions

Patent metadata

Field	Value
Publication number	US-12095838-B2
Application number	US-202117516607-A
Country	US
Kind code	B2
Filing date	Nov 1, 2021
Priority date	Sep 14, 2021
Publication date	Sep 17, 2024
Grant date	Sep 17, 2024

Abstract

In an example system, a meter device records streaming session information. Cluster creation circuitry trains a model by grouping information from multiple streaming sessions into clusters. All streaming sessions within a given cluster have matching media and streaming sources. Model executor circuitry assigns incoming streaming session information to a cluster or to noise. Cluster creation circuitry edits the model by creating new clusters out of information from multiple streaming sessions with similar attributes that were originally labeled as noise.

First claim

What is claimed is:

1. A computing system comprising a processor and a memory, the computing system configured to perform a set of acts comprising:
accessing training data comprising, for each of a plurality of reference television programs, respective reference media information for the reference television program and a respective reference streaming source for the reference television program, wherein the respective reference media information characterizes the reference television program, and wherein the respective reference streaming source identifies a streaming source that provides the reference television program to media devices;
creating, using the training data, a cluster of streaming sessions, the cluster representing streaming sessions where the same television program is provided by a given streaming source;
obtaining meter information indicative of a streaming session detected by a meter, the meter information identifying a detected television program and a detected streaming source;
classifying, using the detected television program and the detected streaming source, the streaming session as belonging to the cluster or as being outside the cluster; and
based on classifying the streaming session as being outside the cluster, identifying the streaming session as an invalid streaming session.

2. The computing system of claim 1 , wherein the cluster has a minimum number of streaming sessions.

3. The computing system of claim 1 , wherein a distance exists between a first streaming session and a second streaming session, the distance determined by at least a title, a name of the streaming sources, and a date and time of presentation for both the first streaming session and second streaming session.

4. The computing system of claim 3 , wherein the distance between an un-assigned streaming session and streaming session in the cluster must be below an epsilon threshold for the un-assigned streaming session to be added to the cluster.

5. The computing system of claim 1 , wherein the set of acts further comprises re-assigning a streaming session labeled as noise to a new cluster.

6. The computing system of claim 1 , wherein the set of acts further comprises:
classifying another streaming session as belonging to the cluster; and
based on classifying the streaming session as belonging to the cluster, identifying the other streaming session as a valid streaming session.

7. A non-transitory machine-readable medium comprising instructions that, when executed, cause a computing system to perform a set of acts comprising:
accessing training data comprising, for each of a plurality of reference television programs, respective reference media information for the reference television program and a respective reference streaming source for the reference television program, wherein the respective reference media information characterizes the reference television program, and wherein the respective reference streaming source identifies a streaming source that provides the reference television program to media devices;
creating, using the training data, a cluster of streaming sessions, the cluster representing streaming sessions where the same television program is provided by a given streaming source;
obtaining meter information indicative of a streaming session detected by a meter, the meter information identifying a detected television program and a detected streaming source;
classifying, using the detected television program and the detected streaming source, the streaming session as belonging to the cluster or as being outside the cluster; and
based on classifying the streaming session as being outside the cluster, identifying the streaming session as an invalid streaming session.

8. The non-transitory machine-readable medium of claim 7 , wherein the cluster has a minimum number of streaming sessions.

9. The non-transitory machine-readable medium of claim 7 , wherein a distance exists between a first streaming session and a second streaming session, the distance determined by at least the title, name of streaming sources, and date and time of presentation for both streaming sessions.

10. The non-transitory machine-readable medium of claim 9 , wherein the distance between an un-assigned streaming session and streaming session in a cluster must be below an epsilon threshold for the un-assigned streaming session to be added to the cluster.

11. The non-transitory machine-readable medium of claim 7 , wherein the set of acts further comprises re-assigning a streaming session labeled as noise to a new cluster.

12. The non-transitory machine-readable medium of claim 11 , wherein the set of acts further comprises:
classifying another streaming session as belonging to the cluster; and
based on classifying the streaming session as belonging to the cluster, identifying the other streaming session as a valid streaming session.

13. A method comprising:
accessing training data comprising, for each of a plurality of reference television programs, respective reference media information for the reference television program and a respective reference streaming source for the reference television program, wherein the respective reference media information characterizes the reference television program, and wherein the respective reference streaming source identifies a streaming source that provides the reference television program to media devices;
creating, using the training data, a cluster of streaming sessions, the cluster representing streaming sessions where the same television program is provided by a given streaming source;
obtaining meter information indicative of a streaming session detected by a meter, the meter information identifying a detected television program and a detected streaming source;
classifying, using the detected television program and the detected streaming source, the streaming session as belonging to the cluster or as being outside the cluster; and
based on classifying the streaming session as being outside the cluster, identifying the streaming session as an invalid streaming session.

14. The method of claim 13 , wherein the cluster has a minimum number of streaming sessions.

15. The method of claim 13 , wherein a distance exists between a first streaming session and a second streaming session, the distance determined by at least the title, name of streaming sources, and date and time of presentation for both streaming sessions.

16. The method of claim 15 , wherein the distance between an un-assigned streaming session and streaming session in the cluster must be below a threshold epsilon for the un-assigned streaming session to be added to the cluster.

17. The method of claim 13 , further including re-assigning a streaming session labeled as noise to a new cluster.

18. The method of claim 17 , further including:
classifying another streaming session as belonging to the cluster; and
based on classifying the streaming session as belonging to the cluster, identifying the other streaming session as a valid streaming session.

Assignees

• Nielsen Co Us Llc

Inventors

• Tapse Sandeep
• Petro James
• Rao Shruthi
• Deshmukh Spoorthi Ramakanth
• Ranganathan Raghuram
• Wright David Howell

Classifications

• G06F18/23

  Clustering techniques · CPC title

• H04L65/1066

  Session management · CPC title

• G06N20/00

  Machine learning · CPC title

• H04L65/80 Primary

  Responding to QoS · CPC title

• H04L65/612 Primary

  for unicast · CPC title