Trend correlations

• US 10,437,910 B2
• Filed: 04/29/2015
• Issued: 10/08/2019
• Est. Priority Date: 04/29/2015
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: entIT Software LLC (Open Text Corporation)
• Inventors: Tzabari, Liron, Benedict, Elad, Levi, Doron
• Primary Examiner(s): Mai, Tan V
• Application Number:
  US15/564,910
• Publication Number:
  US 20180121390A1
• Time in Patent Office: 1,623 Days
• Field of Search: 708490, 708422-426
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: G06F 2218/12 Classification; Matching

Abstract

Examples disclosed herein relate, among other things, to determining a trend correlation. In one aspect, a method is disclosed. The method may include, for example, receiving a first data set associated with a first parameter of an electronic device and a second data set associated with a second parameter of the electronic device. The method may also include generating a first trend set based on the first data set, and generating a second trend set based on the second data set. The method may further include detecting, based on the first trend set and the second trend set, a trend correlation between the first parameter of the electronic device and the second parameter of the electronic device, and providing for display correlation information describing the trend correlation.

Claims

1. A computing device comprising:

a trend detection engine to;

generate a first trend set based on a first data set, wherein the first trend set comprises a first plurality of trend values, wherein each trend value in the first trend set is generated based on a discrete difference between two data values in the first data set, and

generate a second trend set based on a second data set, wherein the second trend set comprises a second plurality of trend values, wherein each trend value in the second trend set is generated based on a discrete difference between two data values in the second data set;

an alignment engine to;

determine a lag amount between the first trend set and the second trend set, and

shift the second trend set by the lag amount to produce a shifted second trend set;

a correlation calculation engine to;

determine a trend correlation between the first data set and the second set by calculating a degree of correlation between the first trend set and the shifted second trend set; and

an output interface to output the trend correlation.

2. The computing device of claim 1, wherein the degree of correlation represents a degree of linear correlation between the first trend set and the shifted second trend set.

3. The computing device of claim 1, wherein the correlation calculation engine is to calculate the degree of correlation by calculating a Pearson product-moment correlation coefficient between the first trend set and the shifted second trend set.

4. The computing device of claim 3, wherein calculating the Pearson product-moment correlation coefficient comprises normalizing the first trend set and the shifted second trend set to cause the first trend set and the shifted second trend set to have zero means.

5. The computing device of claim 1, wherein determining the lag amount comprises shifting the second trend set by a plurality of shift amounts and determining which of the plurality of shift amounts maximizes a dot product of the first trend set and the second trend set.

6. The computing device of claim 1, wherein the output interface is to output the trend correlation by at least one of:

providing the trend correlation for display, storing the trend correlation, and transmitting the trend correlation to another device.

7. The computing device of claim 1, wherein the trend detection engine is further to pre-process at least one of the first data set and the second data set, wherein the pre-processing comprises at least one of formatting and scaling the at least one of the first data set and the second data set.

8. A method comprising:

receiving, by a performance monitoring device, a first data set associated with a first parameter of an electronic device and a second data set associated with a second parameter of the electronic device;

generating, by the performance monitoring device, a first trend set based on the first data set, wherein the first trend set comprises a first plurality of trend values, wherein each trend value in the first trend set is generated based on a discrete difference between two data values in the first data set;

generating a second trend set based on the second data set, wherein the second trend set comprises a second plurality of trend values, wherein each trend value in the second trend set is generated based on a discrete difference between two data values in the second data set;

based on the first trend set and the second trend set, detecting, by the performance monitoring device, a trend correlation between the first parameter of the electronic device and the second parameter of the electronic device; and

providing for display correlation information describing the trend correlation.

9. The method of claim 8, wherein the method further comprises:

determining a time delay between the first trend set and the second trend set; and

shifting at least one of the first trend set and the second trend set by the time delay before detecting the trend correlation.

10. The method of claim 9, wherein determining the time delay comprises shifting the second trend set by a plurality of shift amounts and determining which of the plurality of shift amounts maximizes a dot product of the first trend set and the second trend set.

11. The method of claim 10, wherein detecting the trend correlation comprises calculating a Pearson product-moment correlation coefficient between the first trend set and the shifted second trend set.

12. The method of claim 11, wherein calculating the Pearson product-moment correlation coefficient comprises normalizing the first trend set and the shifted second trend set to cause the first trend set and the shifted second trend set to have zero means.

13. The method of claim 11, comprising:

pre-processing at least one of the first data set and the second data set, wherein the pre-processing comprises at least one of formatting and scaling the at least one of the first data set and the second data set.

14. The method of claim 8, wherein the method further comprises:

determining a time delay between the first data set and the second data set; and

shifting at least one of the first data set and the second data set by the time delay before generating the first trend set and the second trend set.

15. The method of claim 8, wherein generating the first trend set comprises differentiating the first data set, and wherein generating the second trend set comprises differentiating the second data set.

16. A non-transitory machine-readable storage medium encoded with instructions executable by a processor of a computing device to cause the computing device to:

produce a first trend set by differentiating a first data set, wherein the first trend set comprises a first plurality of trend values, wherein each trend value in the first trend set is generated based on a discrete difference between two data values in the first data set;

produce a second trend set by differentiating a second data set, wherein the second trend set comprises a second plurality of trend values, wherein each trend value in the second trend set is generated based on a discrete difference between two data values in the second data set;

produce a shifted second trend set by shifting the second trend set by a delay detected between the first trend set and the second trend set; and

determine a trend correlation between the first data set and the second data set by determining a linear correlation between the first trend set and the shifted second trend set.

17. The non-transitory machine-readable storage medium of claim 16, wherein the instructions are further to cause the computing device to:

detect the delay by shifting the second trend set by a plurality of shift amounts, and determining which of the plurality of shift amounts maximizes a dot product of the first trend set and the second trend set.

18. The non-transitory machine-readable storage medium of claim 16, wherein determining the trend correlation comprises calculating a Pearson product-moment correlation coefficient between the first trend set and the shifted second trend set.

19. The non-transitory machine-readable storage medium of claim 18, wherein calculating the Pearson product-moment correlation coefficient comprises normalizing the first trend set and the shifted second trend set to cause the first trend set and the shifted second trend set to have zero means.

20. The non-transitory machine-readable storage medium of claim 16, wherein the instructions are further to cause the computing device to:

pre-process at least one of the first data set and the second data set, wherein the pre-processing comprises at least one of formatting and scaling the at least one of the first data set and the second data set.