Error correction

Patent metadata

Field	Value
Publication number	US-11237298-B2
Application number	US-201916518205-A
Country	US
Kind code	B2
Filing date	Jul 22, 2019
Priority date	Jul 22, 2019
Publication date	Feb 1, 2022
Grant date	Feb 1, 2022

Abstract

Correction management techniques are provided. In one embodiment, the techniques involve determining, via a first machine learning model, a first and second data based on a respective first and second raw data obtained from a plurality of sensors, determining, based on a deviation between the first data and the second data, an inaccuracy of the first data, identifying an ambient situation corresponding to the first raw data and the second raw data, selecting, from historical raw data of the plurality of sensors, a group of raw data corresponding to the ambient situation, and correcting, via a second machine learning model, the first data based on the selected group of raw data.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
generating, via a first machine learning model, a first and second data based on a respective first raw data obtained from a first sensor and second raw data obtained from a second sensor, wherein the first machine learning model comprises a correction model used to correct the first and second raw data to generate the first and second data, respectively;
determining, based on a deviation between the first data and the second data, an inaccuracy of the first data indicating an inadequacy of the first machine learning model to correct the first raw data;
identifying an ambient situation corresponding to the first raw data and the second raw data;
selecting, from historical raw data of the first and second sensors, a group of raw data corresponding to the ambient situation; and
correcting, via a second machine learning model and after determining the inaccuracy of the first data, the first data based on the selected group of raw data.

2. The method of claim 1 , further comprising:
identifying, by one or more processors, a first ambient attribute of the ambient situation, wherein the first ambient attribute affects the first data more than a second ambient attribute of the ambient situation,
wherein selecting the group of raw data comprises: selecting, by the one or more processors, the group of raw data based on the first ambient attribute.

3. The method of claim 2 , wherein the correcting the first data comprises:
determining, by the one or more processors, a third data based on the selected group of raw data and the first machine learning model;
determining, by the one or more processors, correction data based on the second machine learning model, the selected group of raw data, and the third data; and
correcting, by the one or more processors, the first data based on the first raw data and the correction data.

4. The method of claim 3 ,
wherein the second machine learning model is trained based on the selected group of raw data and the third data, such that the correction data indicates an associative relationship between the selected group of raw data and the third data for the ambient situation.

5. The method of claim 2 , wherein the first ambient attribute comprises any of: a wind direction, a wind level, a temperature, an air pressure, a humidity level, a weather condition, and a geography condition.

6. The method of claim 1 , further comprising:
dividing, by one or more processors, an area occupied by the first sensor and a plurality of sensors into a plurality of regions based on the ambient situation and a spatial relationship between the first sensor and the plurality of sensors, wherein the first raw data is obtained only from the first sensor and the second raw data is obtained from the plurality of sensors, which includes the second sensor, and wherein the group of raw data is selected based on the plurality of regions.

7. The method of claim 6 , wherein selecting the group of raw data based on plurality of regions comprises:
selecting, by one or more processors, a region of the plurality of regions based on the ambient situation and an effect of the region on the first data, wherein the first data is affected by the selected region more than a further region of the plurality of regions, and wherein the group of raw data includes data from a sensor in the selected region.

8. The method of claim 7 , wherein the ambient situation comprises a wind direction, and the selected region is located in an upstream direction of the wind direction.

9. A computer-implemented system, comprising a computer processor coupled to a computer-readable memory unit, the memory unit comprising instructions that when executed by the computer processor implements a method comprising:
generating, via a first machine learning model, a first and second data based on a respective first raw data obtained from a first sensor and second raw data obtained from a second sensor, wherein the first machine learning model comprises a correction model used to correct the first and second raw data to generate the first and second data, respectively;
determining, based on a deviation between the first data and the second data, an inaccuracy of the first data indicating an inadequacy of the first machine learning model to correct the first raw data;
identifying an ambient situation corresponding to the first raw data and the second raw data;
selecting, from historical raw data of the first and second sensors, a group of raw data corresponding to the ambient situation; and
correcting, via a second machine learning model and after determining the inaccuracy of the first data, the first data based on the selected group of raw data.

10. The system of claim 9 , wherein the method further comprises:
identifying a first ambient attribute of the ambient situation, wherein the first ambient attribute affects the first data more than a second ambient attribute of the ambient situation,
wherein selecting the group of raw data comprises: selecting the group of raw data based on the first ambient attribute.

11. The system of claim 10 , wherein the correcting the first data comprises:
determining a third data based on the selected group of raw data and the first machine learning model;
determining correction data based on the second machine learning model, the selected group of raw data, and the third data; and
correcting the first data based on the first raw data and the correction data.

12. The system of claim 11 ,
wherein the second machine learning model is trained based on the selected group of raw data and the third data, such that the correction data indicates an associative relationship between the selected group of raw data and the third data for the ambient situation.

13. The system of claim 10 , wherein the first ambient attribute comprises any of: a wind direction, a wind level, a temperature, an air pressure, a humidity level, a weather condition, and a geography condition.

14. The system of claim 9 , wherein the method further comprises:
dividing an area occupied by the first sensor and a plurality of sensors into a plurality of regions based on the ambient situation and a spatial relationship between the first sensor and the plurality of sensors, wherein the first raw data is obtained only from the first sensor and the second raw data is obtained from the plurality of sensors, which includes the second sensor, and wherein the group of raw data is selected based on the plurality of regions.

15. The system of claim 14 , wherein selecting the group of raw data based on plurality of regions comprises:
selecting a region of the plurality of regions based on the ambient situation and an effect of the region on the first data, wherein the first data is affected by the selected region more than a further region of the plurality of regions, and wherein the group of raw data includes data from a sensor in the selected region.

16. The system of claim 15 , wherein the ambient situation comprises a wind direction, and the selected region is located in an upstream direction of the wind direction.

17. A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by an electronic device to cause the electronic device to perform actions of:
generating, via a first machine learning model, a first and second data based on a respective first raw data obtained from a first sensor and second raw data obtai

Assignees

• IBM

Inventors

• Qu Junmei
• Wang Lingyun
• Xia Xi
• Shao Jin Yan
• Bai Xin Xin

Classifications

• G01W2203/00

  Real-time site-specific personalized weather information, e.g. nowcasting · CPC title

• Y02A90/10

  Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation · CPC title

• G01W1/08

  Adaptations of balloons, missiles, or aircraft for meteorological purposes; Radiosondes (transmission of measurements G01D; transmitting circuits H04B; {balloons in general B64B; aeroplanes in general B64D; missiles in general F42B15/00}) · CPC title

• G01W1/02 Primary

  Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed (G01W1/10 takes precedence) · CPC title

• G06N20/00

  Machine learning · CPC title