CASE-BASED REASONING

• US 20150112916A1
• Filed: 10/17/2013
• Published: 04/23/2015
• Est. Priority Date: 10/17/2013
• Status: Active Grant

• Current Assignee: Baker Hughes Incorporated (Baker Hughes Company)
• Original Assignee: Verdande Technology AS
• Inventors: Gundersen, Odd Erik, Lorentzen, Hkon, Bach, Kerstin, Srmo, Frode
• Granted Patent: US 9,619,765 B2
• Time in Patent Office: Days
• US Class Current: 706/47
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management

Abstract

Disclosed herein is a computer-implemented method of monitoring a situation by generating an overall similarity score between a received data stream and a case in case-based reasoning (CBR) the method comprising: receiving a data stream comprising information on a monitored situation; generating a plurality of parallel data streams, wherein each of the generated plurality of data streams is dependent on the received data stream; generating, for each of the generated data streams, a similarity score for a feature of a case, wherein each similarity score is generated in dependence on a comparison between information in the generated data stream and stored information on the feature of a case, and each of the similarity scores is generated in dependence on a comparison with stored information on a different feature of the same case; and generating an overall similarity score between the received data stream and the case in dependence on the generated similarity scores.

Claims

1. A computer-implemented method of monitoring a situation by generating an overall similarity score between a received data stream and a case in case-based reasoning (CBR) the method comprising:

receiving a data stream comprising information on a monitored situation;

using a computer, generating a plurality of parallel data streams, wherein each of the generated plurality of data streams is dependent on the received data stream;

using a computer, generating, for each of the generated data streams, a similarity score for a feature of a case, wherein each similarity score is generated in dependence on a comparison between information in the generated data stream and stored information on the feature of a case, and each of the similarity scores is generated in dependence on a comparison with stored information on a different feature of the same case; and

generating an overall similarity score between the received data stream and the case in dependence on the generated similarity scores.

2. The method according to claim 1, further comprising generating each similarity score in dependence on comparison information of the corresponding feature of the case.

3. The method according to claim 1, wherein the received data stream comprises one or more streams of parameters and each of the generated data streams comprises one or more streams of parameters in dependence on the received data stream.

4. The method according to claim 1, further comprising generating the similarity score for each feature in dependence on a comparison of the value of a parameter in a generated data stream of parameters for the feature with the stored value of the parameter for the feature.

5. The method according to claim 2, wherein the comparison information comprises weight information and the method further comprises weighting the value of the parameter in the generated data stream for a feature and/or the stored value of the parameter for the feature;

and generating the similarity score of the feature in dependence on the weighted value(s).

6. The method according to claim 2, wherein the comparison information comprises a comparison function that specifies computations for generating the similarity score and the method further comprises generating the similarity score of each feature in dependence on the comparison function of the feature.

7. The method according to claim 1, further comprising generating one or more aggregate similarity scores in dependence on one or more of the generated similarity scores;

and

generating the overall similarity score in dependence on the aggregate similarity score(s).

8. The method according to claim 1, further comprising determining the overall similarity score in real-time.

9. The method according to claim 1, wherein the method is performed by a comparison agent.

10. The method according to claim 9, wherein the comparison agent is configured in dependence on the comparison information for each feature and including the stored information for each feature.

11. The method according to claim 1, wherein the case comprises information that describes a situation in one of the finance industry, healthcare industry or energy industry.

12. A computer-implemented method of monitoring a situation by determining a set of one or more cases in CBR, the method comprising:

receiving a data stream comprising information on a monitored situation;

generating a plurality of parallel data streams from the received data stream;

generating, according to the method of claim 1, an overall similarity score between each of the plurality of parallel data streams and a case, wherein each overall similarity score is generated from a comparison between one of the plurality of parallel data streams and a different one of a plurality of cases; and

determining a set of one or more cases in dependence on the generated overall similarity scores.

13. A non-transitory computer-readable storage medium storing a computer program that, when executed by a computing device, controls the computing device to perform the method of claim 1.