Intelligent learning for explaining anomalies

1. A method for intelligent learning for explaining anomalies by a processor, comprising:

identifying one or more anomalous records identified in a knowledge base;

generating a list of ranked candidate explanations for the one or more anomalous records, wherein each of the ranked candidate explanations are generated according to a knowledge domain, one or more recommendation models, one or more statistical models, one or more machine learning models, one or more rule-based models, one or more reasoning models, or a combination thereof; and

initiating an active learning dialog with one or more users to increase accuracy of the knowledge base, a domain knowledge, and each of the ranked candidate explanations.

2. The method of claim 1 , further including ranking each of the ranked candidate explanations according to a weight and a confidence score.

3. The method of claim 1 , further including ranking each of the ranked candidate explanations according to a level of confidence of evidence in the knowledge base associated with each of the ranked candidate explanations.

4. The method of claim 1 , further including updating the knowledge base using feedback from the one or more users.

5. The method of claim 1 , further including building the one or more recommendation models to generate evidence to support each of the ranked candidate explanations.

6. The method of claim 1 , further including initializing a machine learning mechanism to trigger an active learning operation and to learn, approve, reject, rank, or recommend each of the ranked candidate explanations and evidence in support of each of the ranked candidate explanations, wherein the evidence includes a set of similar cases from other patients or domain experts, domain experts, the domain knowledge, historical data, or a combination thereof.

7. A system for intelligent learning for explaining anomalies, comprising:

one or more computers with executable instructions that when executed cause the system to:

identify one or more anomalous records identified in a knowledge base;

generate list of ranked candidate explanations for the one or more anomalous records, wherein each of the ranked candidate explanations are generated according to a knowledge domain, one or more recommendation models, one or more statistical models, one or more machine learning models, one or more rule-based models, one or more reasoning models or a combination thereof; and

initiate an active learning dialog with one or more users to increase accuracy of the knowledge base, a domain knowledge, and each of the ranked candidate explanations.

8. The system of claim 7 , wherein the executable instructions rank each of the ranked candidate explanations according to a weight and a confidence score.

9. The system of claim 7 , wherein the executable instructions rank each of the ranked candidate explanations according to a level of confidence of evidence in the knowledge base associated with each of the ranked candidate explanations.

10. The system of claim 7 , wherein the executable instructions updating the knowledge base using feedback from the one or more users.

11. The system of claim 7 , wherein the executable instructions build the one or more recommendation models to generate evidence to support each of the ranked candidate explanations.

12. The system of claim 7 , wherein the executable instructions initialize a machine learning mechanism to trigger an active learning operation and to learn, approve, reject, rank, or recommend each of the ranked candidate explanations and evidence in support of each of the ranked candidate explanations, wherein the evidence includes a set of similar cases from other patients or domain experts, domain experts, the domain knowledge, historical data, or a combination thereof.

13. A computer program product for intelligent learning for explaining anomalies by a processor, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

an executable portion that identifies one or more anomalous records identified in a knowledge base;

an executable portion that generates a list of ranked candidate explanations for the one or more anomalous records, wherein each of the ranked candidate explanations are generated according to a knowledge domain, one or more recommendation models, one or more statistical models, one or more machine learning models, one or more rule-based models, one or more reasoning models, or a combination thereof; and

an executable portion that initiate an active learning dialog with one or more users to increase accuracy of the knowledge base, a domain knowledge, and each of the ranked candidate explanations.

14. The computer program product of claim 13 , further including an executable portion that:

ranks each of the ranked candidate explanations according to a weight and a confidence score; and

ranks each of the ranked candidate explanations according to a level of confidence of evidence in the knowledge base associated with each of the ranked candidate explanations.

15. The computer program product of claim 13 , further including an executable portion that updates the knowledge base using feedback from the one or more users.

16. The computer program product of claim 13 , further including an executable portion that builds the one or more recommendation models to generate evidence to support each of the ranked candidate explanations.

17. The computer program product of claim 13 , further including an executable portion that initializes a machine learning mechanism to trigger an active learning operation and to learn, approve, reject, rank, or recommend each of the ranked candidate explanations and evidence in support of each of the ranked candidate explanations, wherein the evidence includes a set of similar cases from other patients or domain experts, domain experts, the domain knowledge, historical data, or a combination thereof.