COMPARING DATA SERIES ASSOCIATED WITH TWO SYSTEMS TO IDENTIFY HIDDEN SIMILARITIES BETWEEN THEM

• US 20110238613A1
• Filed: 03/28/2010
• Published: 09/29/2011
• Est. Priority Date: 03/28/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Shehory, Onn Menahem, Shochat, Moran, Farchi, Eitan Daniel
• Granted Patent: US 8,738,563 B2
• Time in Patent Office: Days
• US Class Current: 706/52
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3664 Environments for testing or debugging software
• CPC Class Codes: G06F 11/3676 for coverage analysis
• CPC Class Codes: G06F 11/3696 Methods or tools to render software testable
• CPC Class Codes: G06F 17/15 Correlation function computation including computation of convolution operations arithmetic circuits for sum of products per se, e.g. multiply-accumulators G06F7/5443; digital filters, e.g. FIR, IIR, adaptive filters H03H17/00
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method that includes: sampling a plurality of parameters associated with a first and a second system to yield a first and a second set of data series, respectively, each data series being associated with samples of a respective parameter; applying, for each two respective data series, a transformation associated with a specified order selected from a plurality of ordered transformations, such that the selected transformation is applied repeatedly to at least one of the two respective data series, wherein, in each repeated application, the transformation is assigned with a different value of a transformation parameter; determining, for each parameter, a one of the plurality of transformation parameters for which the correlation level of the two respective time series is above a specified value; and repeating the applying and the determining with a transformation that is associated with a higher order of the ordered transformations.

Claims

1. A method comprising:

sampling a plurality of parameters associated with a first system and a second system to yield a first set of data series and a second set of data series, respectively, each data series being associated with samples of a respective parameter;

applying, for each two respective data series having a correlation level below a specified threshold, a transformation associated with a specified order selected from a plurality of ordered transformations, such that the selected transformation is applied repeatedly to at least one of the two respective data series, wherein, in each repeated application, the transformation is assigned with a different value of a transformation parameter;

determining, for each parameter, a one of the plurality of transformation parameters for which the correlation level of the two respective data series is above a specified value; and

repeating the applying and the determining with a transformation that is associated with a higher order of the ordered transformations,

wherein at least one of the sampling, the applying, the determining, and the repeating is executed by at least one processor.

2. The method according to claim 1, further comprising associating the applied transformations with respective determined transformation parameters to yield a similarity analysis of respective parameters between the first and the second systems.

3. The method according to claim 1, further comprising analyzing the applied transformations in view of respective determined transformation parameters to yield a set of recommendations being adjustments required for applying to the first system in order to become similar, in view of its parameters, to the second system.

4. The method according to claim 1, wherein the correlation level is determined by computing a Pearson product moment of each two respective data series.

5. The method according to claim 1, wherein at least some of the transformations comprise at least one of:

linear series manipulations;

non-linear series manipulations;

mapping of at least one of the series from time domain into frequency domain.

6. The method according to claim 1, wherein at least one of the data series comprises data points assigned along a unit axis, and wherein at least one of the transformations comprise multiplying the unit axis of least one of the data series with a specified value.

7. The method according to claim 1, wherein at least one data series is a time series associated with a specified time base.

8. The method according to claim 1, further comprising repeatedly reordering the transformations in the repository, in view of the determined transformation parameters, such that computational intensity of the repeated applications is reduced.

9. The method according to claim 1, further comprising repeatedly updating the transformations in the repository, in view of the determined transformation parameters, by adding or removing transformations.

10. A system comprising:

a sampling module configured to sample a plurality of parameters associated with a first system and a second system to yield a first set of data series and a second set of data series, respectively, each data series being associated with samples of a respective parameter;

a repository containing a plurality of ordered transformations; and

a processor configured to;

apply, for each two respective data series having a correlation level below a specified threshold, a transformation associated with a specified order selected from the repository, such that the selected transformation is applied repeatedly to at least one of the two respective data series, wherein in each repeated application, the transformation is assigned with an increased value of a transformation parameter;

determine, for each operational parameter, a one of the plurality of transformation parameters for which the correlation level of the two respective data series is above a specified value; and

repeat the applying and the determining with a transformation that is associated with a higher order of the ordered transformations.

11. The system according to claim 10, wherein the processor is further configured to associate the applied transformations with respective determined transformation parameters to yield a similarity analysis of respective parameters between the first and the second systems.

12. The system according to claim 10, wherein the processor is configured to analyze the applied transformations in view of respective determined transformation parameters to yield a set of recommendations being adjustments required for applying to the first system in order to become similar, in view of its parameters, to the second system.

13. The system according to claim 10, wherein the correlation level is determined by computing a Pearson product moment of each two respective data series.

14. The system according to claim 10, wherein at least some of the transformations comprise at least one of:

linear series manipulations;

non-linear series manipulations; and

mapping of at least one of the series from time domain into frequency domain.

15. The system according to claim 10, wherein the processor is further configured to repeatedly reorder the transformations in the repository, in view the determined transformation parameters, such that computational intensity of the repeated applying is reduced.

16. The system according to claim 10, wherein the processor is further configured to repeatedly update the transformations in the repository, in view the determined transformation parameters, by adding or removing transformations.

17. A computer program product, the computer program product comprising:

a computer readable storage medium having computer readable program embodied therewith, the computer readable program comprising;

computer readable program configured to sample a plurality of parameters associated with a first system and a second system to yield a first set of data series and a second set of data series, respectively, each data series being associated with samples of a respective parameter;

computer readable program configured to apply, for each two respective data series having a correlation level below a specified threshold, a transformation associated with a specified order selected from a plurality of ordered transformations, such that the selected transformation is applied repeatedly to at least one of the two respective data series, wherein, in each repeated application, the transformation is assigned with a different value of a transformation parameter;

computer readable program configured to determine, for each parameter, a one of the plurality of transformation parameters for which the correlation level of the two respective time series is above a specified value; and

computer readable program configured to repeat the applying and the determining with a transformation that is associated with a higher order of the ordered transformations.

18. A method comprising:

presenting a first and a second data series as a first and a second multidimensional graphic representations respectively;

updating values of data points of at least one of the data series, in response to graphical manipulations applied to at least one of the multidimensional graphic representations, to yield updated data series;

determining a quantitative representation of the at least one updated data series or of a relationship between the updated data series; and

repeating the updating and the determining while monitoring the determined quantitative representation vis à

vis respective graphical manipulations,

wherein at least one of the presenting, the updating, the determining, and the repeating is executed by at least one processor.

19. The method according to claim 18, wherein the graphical manipulations comprises at least one of:

moving, selecting, copying, pasting and deleting portions of the graphic representation;

drawing new data points;

erasing data points and portions from the graphic representation;

selecting, annotating, and highlighting at least one data point on the graphic representation; and

holding and moving selected portions to yield stretching or squeezing of the graphic representation.

20. The method according to claim 18, wherein at least some of the graphic manipulations are selected automatically in accordance with a specified order.

21. The method according to claim 18, wherein the monitoring is graphically presented.

22. A system comprising:

a graphical user interface having a display in operative association with an input device; and

a backend module having a processor and a graphic manipulations repository,

wherein the graphical user interface is configured to present over the display, a first and a second data series as a first and a second multidimensional graphic representations respectively,

wherein the processor is configured to;

update values of data points of at least one of the data series, in response to graphical manipulations selected by the input device from the graphic manipulations repository and applied to at least one of the multidimensional graphic representations over the display, to yield updated data series;

determine a quantitative representation of the updated data series or of a relationship between the updated data series; and

repeat the updating and the determining while monitoring the determined correlation level vis à

vis respective graphical manipulations.

23. The system according to claim 22, wherein the input device is at least one of:

mouse, pointing device, touch-screen, touch pad, and keyboard.