Nominal population metric: clustering of nominal application attributes

• US 8,612,176 B2
• Filed: 01/23/2008
• Issued: 12/17/2013
• Est. Priority Date: 01/26/2007
• Status: Active Grant

• Current Assignee: Auburn University
• Original Assignee: Auburn University
• Inventors: Gilbert, Juan E.
• Primary Examiner(s): Raymond, Edward
• Application Number:
  US12/011,053
• Publication Number:
  US 20080183731A1
• Time in Patent Office: 2,155 Days
• Field of Search: 702/81, 702/82, 702/84, 702179-181, 706/60
• US Class Current: 702/179
• CPC Class Codes: G04F 3/00 Apparatus which can be set and started to measure-off predetermined or adjustably-fixed time intervals with driving mechanisms, e.g. dosimeters with clockwork
• CPC Class Codes: G06F 16/2457 with adaptation to user needs
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/2462 Approximate or statistical queries

Abstract

Clustering of nominal attributes using a nominal population metric enables comparisons of entities which are not easily comparable. In some embodiments, nominal population metrics are determined using a similarity matrix and a nominal population matrix using comparisons. In some embodiments, nominal population metrics are determined using a nominal population matrix using distributions. A computing device is able to determine the nominal population metrics with the appropriate hardware and applications configured for computing the nominal population metrics.

Claims

1. A method of determining non-numerical population metrics using a computing device comprising:

a. comparing a plurality of entities with a processor of the computing device thereby forming one or more entity comparisons, wherein each entity has a common non-numerical attribute and corresponding non-numerical value such that the one or more entity comparisons each associate two of the corresponding non-numerical values forming a non-numerical attribute value pair;

b. determining a coverage percentage for one or more of the non-numerical attribute value pairs based on the one or more entity comparisons with the processor;

c. storing the coverage percentage in a memory of the computing device; and

d. determining a non-numerical population metric based on the coverage percentage with the processor;

wherein the two of the corresponding non-numerical values forming the non-numerical attribute value pair come from different entities of the plurality of entities.

2. The method of claim 1 wherein the coverage percentage is determined by an occurrence frequency of each of the one or more of the non-numerical attribute value pairs among all the pairs formed by the entity comparisons for the common non-numerical attribute.

3. The method of claim 1 further comprising determining an inverse coverage percentage by subtracting the coverage percentage from 100%.

4. The method of claim 3 further comprising selecting one or more entities from the plurality of entities based in part on the inverse coverage percentage.

5. The method of claim 1 wherein the one or more non-numerical attributes include at least one of race, gender, major, extra curricular activity, wealth, nationality and country of origin.

6. The method of claim 1 wherein the plurality of entities are selected from the group consisting of admissions applications, job applications and scholarships.

7. The method of claim 1 wherein the attribute value pairs and the coverage percentage are contained in a non-numerical population matrix.

8. The method of claim 1 wherein the computing device is selected from the group consisting of a personal computer, a laptop computer, a computer workstation, a server, a mainframe computer, a handheld computer, a personal digital assistant, a cellular/mobile telephone, a smart appliance and a gaming console.

9. The method of claim 1, wherein one or more difference measures are calculated based on the comparing of the plurality of entities.

10. The method of claim 1, wherein each of the entities is limited to a single non-numerical value for the common non-numerical attribute.

18. The method of claim 9 wherein the difference measures are contained within a similarity matrix.

11. A computing device comprising:

a. a processor;

b. a memory coupled to the processor; and

c. an application stored in the memory and processed by the processor, the application for;

i. comparing a plurality of entities thereby forming one or more entity comparisons, wherein each entity has a common non-numerical attribute and corresponding non-numerical value such that the one or more entity comparisons each associate two of the corresponding non-numerical values forming a non-numerical attribute value pair; and

ii. determining a coverage percentage for one or more of the non-numerical attribute value pairs based on the one or more entity comparisons;

wherein the processor and the memory are integrated into a single device, and further wherein the two of the corresponding non-numerical values forming the non-numerical attribute value pair come from different entities of the plurality of entities.

12. The device of claim 11 wherein the coverage percentage is determined by an occurrence frequency of each of the one or more of the non-numerical attribute value pairs among all the pairs formed by the entity comparisons for the common non-numerical attribute.

13. The device of claim 11 wherein the application further determines an inverse coverage percentage from the coverage percentage by subtracting the coverage percentage from 100%.

14. The device of claim 11, wherein one or more difference measures are calculated based on the comparing of the plurality of entities.

16. The device of claim 11, wherein each of the entities is limited to a single non-numerical value for the common non-numerical attribute.

15. A system comprising:

a. a processing component;

b. a memory component coupled to the processing component; and

c. an application component stored in the memory component and processed by the processing component, the application component for;

i. comparing a plurality of entities thereby forming one or more entity comparisons, wherein each entity has a common non-numerical attribute and corresponding non-numerical value such that the one or more entity comparisons each associate two of the corresponding non-numerical values forming a non-numerical attribute value pair; and

ii. determining a coverage percentage for one or more of the non-numerical attribute value pairs based on the one or more entity comparisons;

wherein the two of the corresponding non-numerical values forming the non-numerical attribute value pair come from different entities of the plurality of entities.

17. The system of claim 15, wherein one or more difference measures are calculated based on the comparing of the plurality of entities.

19. The system of claim 15, wherein each of the entities is limited to a single non-numerical value for the common non-numerical attribute.