Input data correction

Patent metadata

Field	Value
Publication number	US-11119842-B2
Application number	US-201916681761-A
Country	US
Kind code	B2
Filing date	Nov 12, 2019
Priority date	Apr 28, 2017
Publication date	Sep 14, 2021
Grant date	Sep 14, 2021

Abstract

Technical solutions are described that address correcting input time-series data provided for analysis and predictions. An example computer-implemented method includes receiving, by a processor, a time-series data input by a user. The computer-implemented method also includes computing, by the processor, a first plurality of predicted values based on the time-series data input by the user; computing, by the processor, a second plurality of predicted values by. The computer-implemented method also includes determining estimated time-series data based on the time-series data input by the user. The computer-implemented method also includes computing the second plurality of predicted values based on the estimated time-series data. The computer-implemented method also includes determining, by the processor, a defect in the time-series data input by the user based on a distribution of a plurality of differences between respective values from the first plurality of predicted values and the second plurality of predicted values.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving, by a processor, an input time-series data;
computing, by the processor, state variables based on past states in the input time-series data;
computing, by the processor, a first set of predicted values based on the state variables;
computing, by the processor, a second set of predicted values based on a dynamic system model;
determining, by the processor, a distribution of differences between the first set of predicted values and the second set of predicted values;
in response to the distribution of differences not matching a predetermined distribution, notifying, by the processor, that the input time-series data is defective; and
in response to the input time-series data being defective, determining, by the processor, a root cause of a defect in the input time-series data using machine learning.

2. The computer-implemented method of claim 1 , wherein the predetermined distribution is a Gaussian distribution of inaccuracies in the state variable computations.

3. The computer-implemented method of claim 1 , wherein the distribution match is determined by comparing statistical parameters of the distribution of differences and the predetermined distribution, wherein the statistical parameters comprise at least one of a mean and a covariance.

4. The computer-implemented method of claim 1 , further comprising:
displaying, by the processor, a prompt for a user, the prompt displaying the input time-series data and an estimated time-series data to be used instead of the input time-series data; and
receiving, by the processor, a selection of time-series data to be used.

5. The computer-implemented method of claim 4 , further comprising, in response to the selection of the estimated time-series data:
computing, by the processor, a revised predicted value based on the estimated time-series data; and
displaying, by the processor, the revised predicted value.

6. The computer-implemented method of claim 1 , further comprising determining that the cause of the defect in the input time-series data is one from a group of causes comprising under-reporting of the time-series data, over-reporting of the time-series data, and a malfunction of a sensor that provides the time-series data.

7. A system comprising:
a memory; and
a processor coupled with the memory, the processor configured to perform a method comprising:
receiving an input time-series data;
computing state variables based on past states in the input time-series data;
computing a first set of predicted values based on the state variables;
computing a second set of predicted values based on a dynamic system model;
determining a distribution of differences between the first set of predicted values and the second set of predicted values;
in response to the distribution of differences not matching a predetermined distribution, notifying that the input time-series data is defective; and
in response to the input time-series data being defective, determining, by the processor, a root cause of a defect in the input time-series data using machine learning.

8. The system of claim 7 , wherein the predetermined distribution is a Gaussian distribution of inaccuracies in the state variable computations.

9. The system of claim 7 , wherein the distribution match is determined by comparing statistical parameters of the distribution of differences and the predetermined distribution, wherein the statistical parameters comprise at least one of a mean and a covariance.

10. The system of claim 7 , wherein the method further comprises:
displaying, by the processor, a prompt for a user, the prompt displaying the input time-series data and an estimated time-series data to be used instead of the input time-series data; and
receiving, by the processor, a selection of time-series data to be used.

11. The system of claim 10 , wherein the method further comprises, in response to the selection of the estimated time-series data:
computing, by the processor, a revised predicted value based on the estimated time-series data; and
displaying, by the processor, the revised predicted value.

12. The system of claim 7 , further comprising determining that the cause of the defect in the input time-series data is one from a group of causes comprising under-reporting of the time-series data, over-reporting of the time-series data, and a malfunction of a sensor that provides the time-series data.

13. A computer program product comprising a memory storage device having computer executable instructions stored therein, which when executed by a processing unit causes the processing unit to perform a method comprising:
receiving an input time-series data;
computing state variables based on past states in the input time-series data;
computing a first set of predicted values based on the state variables;
computing a second set of predicted values based on a dynamic computer program product model;
determining a distribution of differences between the first set of predicted values and the second set of predicted values;
in response to the distribution of differences not matching a predetermined distribution, notifying that the input time-series data is defective; and
in response to the input time-series data being defective, determining, by the processor, a root cause of a defect in the input time-series data using machine learning.

14. The computer program product of claim 13 , wherein the predetermined distribution is a Gaussian distribution of inaccuracies in the state variable computations.

15. The computer program product of claim 13 , wherein the distribution match is determined by comparing statistical parameters of the distribution of differences and the predetermined distribution, wherein the statistical parameters comprise at least one of a mean and a covariance.

16. The computer program product of claim 13 , wherein the method further comprises:
displaying, by the processor, a prompt for a user, the prompt displaying the input time-series data and an estimated time-series data to be used instead of the input time-series data; and
receiving, by the processor, a selection of time-series data to be used.

17. The computer program product of claim 16 , wherein the method further comprises, in response to the selection of the estimated time-series data:
computing a revised predicted value based on the estimated time-series data; and displaying the revised predicted value.

Assignees

• IBM

Inventors

• Chang Hung-Yang
• Codella James V
• Das Subhro

Classifications

• G06F11/079 Primary

  Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36) · CPC title

• G05B23/024

  Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06F11/3024

  where the computing system component is a central processing unit [CPU] · CPC title

• G06F11/0742

  in a data processing system embedded in a mobile device, e.g. mobile phones, handheld devices · CPC title