System and method for temporal data mining

• US 7,526,461 B2
• Filed: 08/09/2005
• Issued: 04/28/2009
• Est. Priority Date: 11/17/2004
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (Motors Liquidation Co. GUC Trust)
• Inventors: Barajas, Leandro G., Srinivasa, Narayan
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Bharadwaj; Kalpana
• Application Number:
  US11/199,698
• Publication Number:
  US 20060106797A1
• Time in Patent Office: 1,358 Days
• Field of Search: 706/20, 706/21, 706/48, 707/102, 715/513
• US Class Current: 706/20
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/2477 Temporal data queries
• CPC Class Codes: G06N 3/086 using evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A system, method, and apparatus for signal characterization, estimation, and prediction comprising an integrated search algorithm that cooperatively optimizes several data mining sub-tasks, the integrated search algorithm including a machine learning model, and the method comprising processing the data for data embedding, data embedding the processed data for searching for patterns, extracting time and frequency patterns, and training the model to represent learned patterns for signal characterization, estimation, and prediction.

Claims

1. A computer implemented method of signal characterization for event code analysis in a manufacturing facility, comprising an integrated search algorithm that cooperatively optimizes data mining sub-tasks, the integrated search algorithm including a machine learning model, the integrated search algorithm comprising:

processing the data for data embedding that includes extracting quantity or quality features from the data;

data embedding the processed data for searching for patterns;

extracting time and frequency patterns to provide training samples that include temporal and useful patterns of data that can be used for modeling, estimation, prediction and/or analysis of data features; and

training the machine learning model to represent the temporal and useful data patterns for signal characterization according to the training samples, wherein the temporal and useful patterns of data include information that is used to optimize a system or process.

2. The method of claim 1, wherein the integrated search algorithm cooperatively optimizes data mining tasks concurrently.

3. The method of claim 1, wherein the integrated search algorithm cooperatively optimizes data mining tasks hierarchically.

4. The method as recited in claim 1 wherein processing the data for data embedding comprises utilizing an automated clustering algorithm.

5. The method of claim 1, wherein processing the data for data embedding comprises feature extraction.

6. The method of claim 1, wherein the integrated search algorithm comprises a genetic algorithm.

7. The method of claim 1, wherein extracting time and frequency patterns comprises temporal data mining.

8. The method of claim 1, wherein the machine learning model comprises a neural network.

9. A system for signal characterization for event code analysis in a manufacturing facility, comprising:

a processor to execute instructions of a software module;

an instruction module comprising an integrated search algorithm that cooperatively optimizes data mining sub-tasks, the integrated search algorithm including a machine learning model;

a processing module for processing the data for data embedding that includes extracting quantity or quality features from the data that can be used for modeling, estimation, prediction and/or analysis of data features;

an embedding module for data embedding the processed data for searching for time and frequency patterns;

an extraction module for extracting time and frequency patterns to provide training samples that include temporal and useful patterns of data that can be used for modeling, estimation, prediction and/or analysis of data features; and

a training module for training the machine learning model to represent the temporal and useful data patterns according to the training samples, wherein the temporal and useful data patterns include information that is used to optimize a system or process.

10. The system of claim 9, wherein the integrated search algorithm cooperatively optimizes data mining tasks simultaneously.

11. The system of claim 9, wherein the integrated search algorithm cooperatively optimizes data mining tasks hierarchically.

12. The system of claim 9, wherein the integrated search algorithm comprises an optimization routine.

13. The system of claim 9, wherein the integrated search algorithm comprises a genetic algorithm.

14. The system of claim 9, wherein data embedding the processed data comprises feature extraction.

15. The system as recited in claim 9 wherein data embedding the processed data comprises clustering.

16. The system of claim 9, wherein extracting time and frequency patterns comprises temporal data mining.

17. The system of claim 9, wherein the machine learning model comprises a neural network.

18. A method for temporal data mining utilizing a processor to execute instructions of a software module and a database comprising time series data;

executing genetic algorithm instructions, comprising;

a binning and classifying module;

a data embedding module that extracts quantity or quality features from the data;

a temporal pattern extraction module that extracts samples including temporal and useful patterns of data that can be used for modeling, estimation, prediction and/or analysis of data features; and

a neural network training module;

processing the temporal and useful patterns of data by the genetic algorithm instructions by training the neural network training module to represent the temporal and useful data patterns; and

Generating prediction output, wherein the temporal and useful patterns include information that is used to optimize a manufacturing system or process.

19. The system of claim 18 wherein the genetic algorithm includes evaluation of a fitness function to provide a fitness value, and propagates to a set of next generation chromosomes with higher fitness values over a set of chromosomes with lower fitness values.

20. The method of claim 18 wherein a neural network trained in the neural network training module predicts a downtime, frequency, or mean time to repair event of a machine station in a production facility.