Methods and apparatus to virtually estimate cardinality with global registers

Patent metadata

Field	Value
Publication number	US-12229098-B2
Application number	US-202218148337-A
Country	US
Kind code	B2
Filing date	Dec 29, 2022
Priority date	Oct 28, 2022
Publication date	Feb 18, 2025
Grant date	Feb 18, 2025

Abstract

Methods, apparatus, systems, and articles of manufacture to virtually estimate cardinality with global registers are disclosed. An example apparatus includes processor circuitry to s assign subsets of a sample dataset to a shared global register array, the shared global register array having a first number of registers, the sample dataset selected from a reference dataset of media assets; identify a virtual register array from the shared global register array that includes data elements associated with a label value, the virtual register array including a second number of registers less than the first number of registers; determine a maximum rank value of the label value across the virtual register array; and calculate a cardinality estimate of the label value across the virtual register array based on the second number of registers and the maximum rank value.

First claim

What is claimed is:

1. An apparatus to virtually estimate cardinality with global registers, the apparatus comprising:
at least one memory;
machine readable instructions; and
one or more processors configured to execute the machine readable instructions to at least:
assign subsets of a sample dataset to a shared global register array, the shared global register array having a first number of registers, the sample dataset selected from a reference dataset of media assets;
identify a virtual register array from the shared global register array that includes data elements associated with a label value, the virtual register array including a second number of registers less than the first number of registers;
determine a maximum rank value of the label value across the virtual register array; and
calculate a cardinality estimate of the label value across the virtual register array based on the second number of registers and the maximum rank value.

2. The apparatus of claim 1 , wherein the cardinality estimate is a first cardinality estimate, and the one or more processors are configured to calculate a second cardinality estimate of the label value across the shared global register array based on the first cardinality estimate, the first number of registers, and the second number of registers.

3. The apparatus of claim 1 , wherein the one or more processors are configured to generate a first rank distribution array for the shared global register array.

4. The apparatus of claim 3 , wherein the one or more processors are configured to generate a second rank distribution array for the virtual register array.

5. The apparatus of claim 4 , wherein the one or more processors are configured to generate an estimated recovered rank distribution array for the label value based on the first rank distribution array and the second rank distribution array.

6. The apparatus of claim 5 , wherein the one or more processors are configured to determine an estimated cumulative distribution function for the label value based on the estimated recovered rank distribution array.

7. The apparatus of claim 6 , wherein the one or more processors are configured to determine the maximum rank value for the label value based on the estimated cumulative distribution function.

8. At least one non-transitory machine readable storage medium comprising instructions that, when executed, cause one or more processors to at least
assign subsets of a sample dataset to a shared global register array, the shared global register array having a first number of registers, the sample dataset selected from a reference dataset of media assets;
identify a virtual register array from the shared global register array that includes data elements associated with a label value, the virtual register array including a second number of registers less than the first number of registers;
determine a maximum rank value of the label value across the virtual register array; and
calculate a cardinality estimate of the label value across the virtual register array based on the second number of registers and the maximum rank value.

9. The at least one non-transitory machine readable storage medium of claim 8 , wherein the cardinality estimate is a first cardinality estimation, and the instructions cause the one or more processors to calculate a second cardinality estimate of the label value across the shared global register array based on the first cardinality estimate, the first number of registers, and the second number of registers.

10. The at least one non-transitory machine readable storage medium of claim 8 , wherein the instructions cause the one or more processors to generate a first rank distribution array for the shared global register array.

11. The at least one non-transitory machine readable storage medium of claim 10 , wherein the instructions cause the one or more processors to generate a second rank distribution array for the virtual register array.

12. The at least one non-transitory machine readable storage medium of claim 11 , wherein the instructions cause the one or more processors to generate an estimated recovered rank distribution array for the label value based on the first rank distribution array and the second rank distribution array.

13. The at least one non-transitory machine readable storage medium of claim 12 , wherein the instructions cause the one or more processors to determine an estimated cumulative distribution function for the label value based on the estimated recovered rank distribution array.

14. The at least one non-transitory machine readable storage medium of claim 13 , wherein the instructions cause the one or more processors to determine the maximum rank value for the label value based on the estimated cumulative distribution function.

15. A method to virtually estimate cardinality with global registers, the method comprising:
assigning subsets of a sample dataset to a shared global register array, the shared global register array having a first number of registers, the sample dataset selected from a reference dataset of media assets;
identifying a virtual register array from the shared global register array that includes data elements associated with a label value, the virtual register array including a second number of registers less than the first number of registers;
determining a maximum rank value of the label value across the virtual register array; and
calculating a cardinality estimate of the label value across the virtual register array based on the second number of registers and the maximum rank value.

16. The method of claim 15 , wherein the cardinality estimate is a first cardinality estimate, further including calculating a second cardinality estimate of the label value across the shared global register array based on the first cardinality estimate, the first number of registers, and the second number of registers.

17. The method of claim 15 , further including generating a first rank distribution array for the shared global register array.

18. The method of claim 17 , further including generating a second rank distribution array for the virtual register array.

19. The method of claim 18 , further including generating an estimated recovered rank distribution array for the label value based on the first rank distribution array and the second rank distribution array.

20. The method of claim 19 , further including determining an estimated cumulative distribution function for the label value based on the estimated recovered rank distribution array.

21. The method of claim 20 , further including determining the maximum rank value for the label value based on the estimated cumulative distribution function.

Assignees

• Nielsen Co Us Llc

Inventors

• Sheppard Michael R

Classifications

• G06F16/24556

  Aggregation; Duplicate elimination · CPC title

• G06F16/2228 Primary

  Indexing structures · CPC title