Methods and apparatus to estimate audience sizes of media using deduplication based on multiple vectors of counts

Patent metadata

Field	Value
Publication number	US-12032535-B2
Application number	US-202016917459-A
Country	US
Kind code	B2
Filing date	Jun 30, 2020
Priority date	Jun 30, 2020
Publication date	Jul 9, 2024
Grant date	Jul 9, 2024

Abstract

Disclosed examples to estimate audience sizes of media include a coefficient generator to determine coefficient values for a polynomial based on normalized weighted sums of variances, a normalized weighted sum of covariances, and cardinalities corresponding to a first plurality of vectors of counts from a first database proprietor and a second plurality of vectors of counts from a second database proprietor, a real roots solver to determine a real root value of the polynomial, the real root value indicative of a number of audience members represented in the first plurality of vectors of counts that are also represented in the second plurality of vectors of counts, and an audience size generator to determine the audience size based on the real root value and the cardinalities of the first plurality of vectors of counts and the second plurality of vectors of counts.

First claim

What is claimed is:

1. An apparatus to determine an audience size for media based on sketch data, the apparatus comprising:
memory;
programmable circuitry; and
instructions to cause the programmable circuitry to:
obtain (a) first sketch data included in a first network communication from a first server of a first database proprietor and (b) second sketch data included in a second network communication from a second server of a second database proprietor, the first sketch data to represent a first audience of a media item including first subscribers of the first database proprietor, the second sketch data to represent a second audience of the media item including second subscribers of the second database proprietor, the first sketch data including a first plurality of vectors of counts, each vector of counts of the first plurality of vectors of counts representing the first audience as a distribution of first hash values, the first hash values corresponding to the first subscribers and generated using a hash algorithm, each vector of counts of the first plurality of vectors of counts including a plurality of bins having respective bin numbers, the first hash values usable to determine bin numbers for assigning the first subscribers to respective bins of the plurality of bins, the second sketch data including a second plurality of vectors of counts, each vector of counts of the second plurality of vectors of counts representing the second audience as a distribution of second hash values, the second hash values corresponding to the second subscribers and generated using the hash algorithm, each vector of counts of the second plurality of vectors of counts including the plurality of bins, the second hash values usable to determine bin numbers for assigning the second subscribers to respective bins of the plurality of bins, the first and second sketch data to preserve privacy of the first and second subscribers by representing the first and second subscribers via the first and second pluralities of vectors of counts instead of sharing identifying information associated with the first and second subscribers, wherein the first audience is known to the first database proprietor but not to the second database proprietor, and wherein the second audience is known to the second database proprietor but not to the first database proprietor;
determine coefficient values for a polynomial based on (a) normalized weighted sums of variances, (b) a normalized weighted sum of covariances, and (c) cardinalities, the normalized weighted sums of variances, the normalized weighted sum of covariances, and the cardinalities corresponding to the first plurality of vectors of counts and the second plurality of vectors of counts;
determine a real root value of the polynomial, the real root value indicative of a number of audience members of the first audience represented in the first plurality of vectors of counts that are also represented in the second plurality of vectors of counts;
determine the audience size based on the real root value and the cardinalities of the first plurality of vectors of counts and the second plurality of vectors of counts; and
cause transmission of the audience size to a customer computer.

2. The apparatus of claim 1 , wherein the instructions are to cause the programmable circuitry to determine the normalized weighted sums of variances and the normalized weighted sum of covariances corresponding to the first plurality of vectors of counts and the second plurality of vectors of counts.

3. The apparatus of claim 2 , wherein the instructions are to cause the programmable circuitry to determine the normalized weighted sums of variances and the normalized weighted sum of covariances by:
generating a plurality of first variances based on the first plurality of vectors of counts;
generating a plurality of second variances based on the second plurality of vectors of counts; and
generating a plurality of covariances based on the first plurality of vectors of counts and the second plurality of vectors of counts.

4. The apparatus of claim 3 , wherein the instructions are to cause the programmable circuitry to determine the normalized weighted sums of variances and the normalized weighted sum of covariances by:
generating a first weighted sum of variances based on the plurality of first variances, the first weighted sum of variances corresponding to a first normalized weighted sum of variances of the first plurality of vectors of counts;
generating a second weighted sum of variances based on the plurality of second variances, the second weighted sum of variances corresponding to a second normalized weighted sum of variances of the second plurality of vectors of counts; and
generating a weighted sum of covariances based on the plurality of covariances, the weighted sum of covariances corresponding to the normalized weighted sum of covariances of the first plurality of vectors of counts and the second plurality of vectors of counts.

5. The apparatus of claim 4 , wherein the instructions are to cause the programmable circuitry to generate the first weighted sum of variances, the second weighted sum of variances, and the weighted sum of covariances based on a probability factor corresponding to lengths of the first plurality of vectors of counts and the second plurality of vectors of counts.

6. The apparatus of claim 1 , wherein the instructions are to cause the programmable circuitry to zero-mean data in at least one of the first plurality of vectors of counts or the second plurality of vectors of counts before the programmable circuitry is to determine the coefficient values.

7. The apparatus of claim 1 , wherein the instructions are to cause the programmable circuitry to select equations to be used by the programmable circuitry in response to obtaining a noise term from at least one of the first database proprietor or the second database proprietor.

8. A method to determine an audience size for media based on sketch data, the method comprising:
obtaining (a) first sketch data included in a first network communication from a first server of a first database proprietor and (b) second sketch data included in a second network communication from a second server of a second database proprietor, the first sketch data to represent a first audience of a media item including first subscribers of the first database proprietor, the second sketch data to represent a second audience of the media item including second subscribers of the second database proprietor, the first sketch data including a first plurality of vectors of counts, each vector of counts of the first plurality of vectors of counts representing the first audience as a distribution of first hash values, the first hash values corresponding to the first subscribers and generated using a hash algorithm, each vector of counts of the first plurality of vectors of counts including a plurality of bins having respective bin numbers, the first hash values usable to determine bin numbers for assigning the first subscribers to respective bins of the plurality of bins, the second sketch data including a second plurality of vectors of counts, each vector of counts of the second plurality of vectors of counts representing the second audience as a distribution of second hash values, the second hash values corresponding to the second subscribers and generated using the hash algorithm, each vector of counts of the second plurality of vectors of counts including the plurality of bins, the second hash values usable to determine bin numbers for assigning the second subscribers to respective bins of the plurality of bins, the first and second sketch data to preserve privacy of the first and second subscribers by representing the first and s

Assignees

• Nielsen Co Us Llc

Inventors

• Sheppard Michael R
• Dailey Jake Ryan
• Forthomme Damien
• Sullivan Jonathan
• Brinson Jessica
• Summers Christie Nicole
• Morovati Lopez Diane
• Poppie Molly

Classifications

• H04N21/4667

  Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections · CPC title

• G06Q30/0205 Primary

  based on location or geographical consideration · CPC title

• G06Q50/265

  Personal security, identity or safety · CPC title

• H04N21/835

  Generation of protective data, e.g. certificates {(protecting software against unauthorised usage in a vending or licensing environment G06F21/10)} · CPC title

• H04N21/2408

  Monitoring of the upstream path of the transmission network, e.g. client requests (monitoring data switching networks utilization H04L43/0876; scheduling or organising the servicing of application requests H04L67/60) · CPC title