Method and system to compare data objects

• US 8,180,777 B2
• Filed: 10/24/2010
• Issued: 05/15/2012
• Est. Priority Date: 01/17/2007
• Status: Active Grant

• Current Assignee: Aptima Inc.
• Original Assignee: Aptima Inc.
• Inventors: Duchon, Andrew, Kelton, Kari, Foster, Pacey, Orvis, Kara, McCormack, Robert
• Primary Examiner(s): Breene, John E
• Assistant Examiner(s): Ly, Anh
• Application Number:
  US12/910,823
• Publication Number:
  US 20110040764A1
• Time in Patent Office: 569 Days
• Field of Search: 707/738, 707/739, 707/771, 707/794, 707/E17.089, 707/E17.098, 706/45, 706/46, 706/52, 706/54
• US Class Current: 707/738
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

The present invention relates in general to methods and systems for comparing and maximizing the optimal selection of a first set of one or more data objects to a set of second data objects. In one embodiment, the first set of data objects represent one or more tasks to be fulfilled by a set of capabilities represented by the second data objects. In one embodiment, methods and systems are provided that apply topic modeling and similarity metrics to determine the optimal selection. In one embodiment, methods and systems are provided to determine the appropriateness of a set of second data objects to satisfy the requirements of a first data object given interaction attributes. Embodiments may be used to compare mission requirements with potential team members to determine the appropriateness of team members and teams for a given mission based on interaction attributes of the team members and teams.

Claims

1. A processor-based method for comparing a plurality of data objects comprising the steps of:

receiving a first data object;

receiving a plurality of second data objects;

applying a topic model technique to the first data object and the plurality of second data objects creating a topic model;

grouping the plurality of second data objects creating a plurality of groupings;

the plurality of groupings comprising possible sets of second data objects;

determining a similarity of the first data object and the plurality of groupings;

determining an optimal selection of the first data object and the plurality of groupings from the similarity; and

communicating the optimal selection.

2. The processor-based method of claim 1 wherein the topic model technique comprises Latent Dirichlet Allocation (LDA).

3. The processor-based method of claim 1 wherein the first data object comprises at least one requirement and the plurality of second data objects comprises a plurality of capabilities of a plurality of individuals.

4. The processor-based method of claim 1 wherein the step of determining the optimal selection of the first data object and the plurality of groupings comprises determining a minimal spanning set of the second data objects to maximize the similarity of the first data object and the plurality of groupings.

5. The processor-based method of claim 1 further comprising:

determining an appropriateness of the first data objects and the second data objects; and

determining an optimal selection from the similarity and the appropriateness.

6. The processor-based method of claim 5 further comprising:

receiving at least one third data object;

the step of applying the topic model technique further comprises pre-processing the first data object, the plurality of second data object and the at least one third data object whereby the data objects are normalized; and

applying the topic model technique to the first data object, the plurality of second data objects and the at least one third data object to create the topic model.

7. A processor-based method for comparing a plurality of data objects comprising the steps of:

receiving a first data object;

receiving a plurality of second data objects;

applying a topic model technique to the first data object and the plurality of second data objects creating a topic model;

grouping the plurality of second data objects creating a plurality of groupings;

the plurality of groupings comprising possible sets of second data objects;

determining a first similarity of the first data object and the plurality of groupings;

determining a second similarity of the plurality of second data objects in each of the plurality of groupings;

comparing the first similarity and the second similarity to determine an optimal selection of the first data object and the plurality of groupings; and

communicating the optimal selection.

8. The processor-based method of claim 7 wherein the topic model technique comprises Latent Dirichlet Allocation (LDA).

9. The processor-based method of claim 7 wherein the first data object comprises at least one requirement and the plurality of second data objects comprises a plurality of capabilities of a plurality of individuals.

10. The processor-based method of claim 9 further comprising:

determining an appropriateness for the first data objects and the second data objects; and

the step of comparing the first similarity and the second similarity to determine an optimal selection comprises comparing the first similarity, the second similarity and the appropriateness to determine the optimal selection.

11. The processor-based method of claim 10 further comprising:

receiving at least one third data object;

the step of applying the topic model technique further comprises pre-processing the first data object, the plurality of second data objects and the at least one third data object whereby the data objects are normalized; and

applying the topic model technique to the first data object, the plurality of second data objects and the at least one third data object to create the topic model.

12. A processor-based system for comparing a plurality of data objects comprising the steps of:

means for receiving a first data object;

means for receiving a plurality of second data objects;

means for applying a topic model technique to the first data object and the plurality of second data objects creating a topic model;

means for grouping the plurality of second data objects creating a plurality of groupings;

the plurality of groupings comprising possible sets of second data objects;

means for determining a similarity of the first data object and the plurality of groupings;

means for determining an optimal selection of the first data object and the plurality of groupings from the similarity; and

means for communicating the optimal selection.

13. The system of claim 12 wherein:

the means for receiving a first data object and plurality of second data objects comprises a processor;

the means for applying a topic model technique, the means for grouping the plurality of second data objects, the means for determining a similarity and the means for determining an optimal selection comprises a computer program product in a memory for execution by the processor; and

the means for communicating the optimal selection comprises the processor in communication with a system bus.

14. A computer readable medium having stored thereon a computer program product that, when executed, causes a processor based computer to perform the steps of:

receiving a first data object;

receiving a plurality of second data objects;

applying a topic model technique to the first data object and the plurality of second data objects creating a topic model;

grouping the plurality of second data objects creating a plurality of groupings;

the plurality of groupings comprising possible sets of second data objects;

determining a similarity of the first data object and the plurality of groupings;

determining an optimal selection of the first data object and the plurality of groupings; and

communicating the optimal selection.

15. A processor-based method for comparing a plurality of data object comprising the steps of:

receiving a first data object;

receiving a plurality of second data objects;

applying an interaction measure to the first data object and the plurality of second data objects;

grouping the plurality of second data objects creating a plurality of groupings;

the plurality of groupings comprising possible sets of second data objects;

determining an appropriateness of the first data object and the plurality of groupings;

determining an optimal selection of the first data object and the plurality of groupings from the appropriateness; and

communicating the optimal selection.

16. The processor-based method of claim 15 wherein the plurality of second data objects comprises a plurality of attributes of a plurality of individuals.

17. The processor-based method of claim 15 wherein the plurality of second data objects comprises a plurality of attributes of a plurality of individuals and at least one of the plurality of attributes comprises an interaction attribute.

18. The processor-based method of claim 15 wherein:

the step of determining the optimal selection comprises determining the optimal selection from a group appropriateness and a mutual appropriateness.

19. The processor-based method of claim 15 wherein:

the step of determining the appropriateness of the first data object and the plurality of groupings comprises applying a first appropriateness metric to determine a group appropriateness of the first data object and the plurality of groupings and applying a second appropriateness metric to determine a mutual appropriateness of the second data objects and the plurality of groupings; and

the step of determining the optimal selection comprises determining the optimal selection from the group appropriateness and the mutual appropriateness.