Energy minimization for classification, pattern recognition, sensor fusion, data compression, network reconstruction and signal processing

• US 6,993,186 B1
• Filed: 12/23/1998
• Issued: 01/31/2006
• Est. Priority Date: 12/29/1997
• Status: Expired due to Term

• Current Assignee: Abel Wolman, Jeff B. Glickman
• Original Assignee: Abel Wolman, Jeff B. Glickman
• Inventors: Glickman, Jeff B., Wolman, Abel
• Primary Examiner(s): Mehta, Bhavesh M.
• Assistant Examiner(s): DESIRE, GREGORY M
• Application Number:
  US09/581,949
• Time in Patent Office: 2,596 Days
• Field of Search: 382/133, 382/170, 382/224, 382/281, 382155-160, 382260-266, 382276-280, 706/20
• US Class Current: 382/181
• CPC Class Codes: G06F 18/2137 based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

An analyzer/classifier tool (100) for data comprises use of an energy minimization process (120) with data transformed by an input process (110) into one or more input matrices. The data to be analyzed/classified is processed by an energy minimization technique such as individual differences multidimensional scaling (IDMDS) to produce at least a rate of change of stress/energy. Using the rate of change of stress/energy and possibly other IDMDS output, a back end process (130) analyzes and classifies data through patterns recognized within the data.

Claims

1. A method for classifying data, the method comprising the steps of:

receiving input data for classification;

defining one or more transformations of the input data;

applying energy minimization to the one or more transforms of the input data;

producing a rate of change in energy in response to energy minimization; and

classifying the input data using the rate of change in energy.

2. The method of claim 1 wherein the step of applying energy minimization comprises using individual differences multidimensional scaling applied to the input data.

3. The method of claim 1 wherein the step of applying energy minimization comprises using a finite element method analysis applied to the input data.

4. The method of claim 1 wherein the step of applying energy minimization comprises using simulated annealing applied to the input data.

5. The classification method of claim 1 further comprising the steps of:

producing a change in energy in response to energy minimization; and

classifying the input data using the change in energy.

6. The classification method of claim 1 further comprising the steps of:

producing a minimum energy in response to energy minimization; and

classifying the input data using the minimum energy.

7. A method for classifying data, the method comprising the steps of:

receiving input data for classification;

creating a plurality of data structures from the received input data;

applying least energy merging to the plurality of data structures;

producing at least a change in energy in response to the least energy merging of the plurality of data structures; and

classifying the input data using at least the change in energy.

8. The method of claim 7 wherein the step of applying least energy merging to the plurality or data structures comprises using individual differences multidimensional scaling applied to the plurality of data structures.

9. The method of claim 7 wherein the step of applying least energy merging to the plurality of data structures comprises using a finite element method analysis.

10. The method of claim 7 wherein the step of applying least energy merging to the plurality of data structures comprises using simulated annealing.

11. The method of claim 7 further comprising the steps of:

producing relative deformations from the least energy merged plurality of data structures; and

using the relative deformations to classify the input data.

12. The method of claim 11 further comprising the steps of

producing a classification space output from the relative deformations of the least energy merged plurality of data structures;

and

classifying the input data using the classification space output.

13. The method of claim 7 further comprising the steps of:

producing a least energy merged configuration from the least energy merging of the plurality of data structures;

producing relative deformations of the plurality of data structures from the least energy merged configuration; and

classifying the input data using the relative deformations from the least energy merged configuration.