Intelligent evidence classification and notification in a deep question answering system
System, method, and computer program product to identify changes in evidence used to answer questions by a deep question answering system, by identifying a first evidence related to a feature, the deep question answering system having identified the feature as being relevant to answering a first type of question of a plurality of types of questions, and responsive to: (i) detecting a change in the first evidence, and (ii) determining that a confidence score of a first response generated for a first question, of the first type of question, exceeds a confidence threshold, generating an updated response for the first question based on the changed first evidence.
1. A computer program product to identify changes in evidence used to answer questions by a question answering system, the computer program product comprising:
a non-transitory computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code comprising:
computer-readable program code configured to identify a first evidence related to a feature, wherein the question answering system has identified the feature as being relevant to answering a first type of question of a plurality of types of questions;
computer-readable program code configured to detect a change in the first evidence;
computer-readable program code configured to determine that a confidence score of a first response generated for a first question, of the first type of question, exceeds a confidence threshold;
computer readable program code configured to output, to a user, a notification of the change in the first evidence; and
computer readable program code configured to, responsive to receiving approval from the user to update the first response, generate an updated response for the first question based on the change in the first evidence and the confidence score exceeding the confidence threshold.
2. The computer program product of claim 1 , wherein a user provides the approval.
3. The computer program product of claim 1 , wherein the updated response is based on at least one of: (i) processing of a second question, of the first type, and (ii) reprocessing the first question.
4. The computer program product of claim 3 , wherein the second question is generated by one of: (i) the question answering system, and (ii) a user.
5. The computer program product of claim 1 , wherein the first evidence is unavailable, wherein the change in the first evidence comprises an availability of the first evidence.
6. The computer program product of claim 1 , wherein the first evidence is available, wherein the change in the first evidence comprises a change in a content of the first evidence.
7. The computer program product of claim 1 , further comprising monitoring the first evidence to detect the change.
8. The computer program product of claim 7 , wherein the first evidence is monitored upon determining that a relevance score of the first evidence exceeds a relevance threshold, wherein the first evidence comprises a content of at least one of: (i) a document, (ii) a variable, (iii) an article, and (iv) a web page, and wherein the first question is part of a first case comprising at least the first question.
9. A system, comprising:
one or more computer processors; and
a memory containing a program which when executed by the one or more computer processors, performs an operation to identify changes in evidence used to answer questions by a deep question answering system, the operation comprising:
identifying a first evidence related to a feature, wherein the question answering system has identified the feature as being relevant to answering a first type of question of a plurality of types of questions;
detecting a change in the first evidence
determining that a confidence score of a first response generated for a first question, of the first type of question, exceeds a confidence threshold;
outputting, to a user, an indication of the change in the first evidence; and
responsive to receiving approval from the user to update the first response, generating an updated response for the first question based on the change in the first evidence and the confidence score exceeding the confidence threshold.
10. The system of claim 9 , wherein a user provides the approval.
11. The system of claim 9 , wherein the updated response is based on at least one of: (i) processing of a second question, of the first type, and (ii) reprocessing the first question.
12. The system of claim 11 , wherein the second question is generated by one of: (i) the question answering system, and (ii) a user.
13. The system of claim 9 , wherein the first evidence is unavailable, wherein the change in the first evidence comprises an availability of the first evidence.
14. The system of claim 9 , wherein the first evidence is available, wherein the change in the first evidence comprises a change in a content of the first evidence.
15. The system of claim 9 , further comprising monitoring the first evidence to detect the change.
16. The system of claim 15 , wherein the first evidence is monitored upon determining that a relevance score of the first evidence exceeds a relevance threshold, wherein the first evidence comprises a content of at least one of: (i) a document, (ii) a variable, (iii) an article, and (iv) a web page, and wherein the first question is part of a first case comprising at least the first question.
17. A non-transitory computer readable storage medium storing instructions which when executed by a processor performs an operation to identify changes in evidence used to answer questions by a question answering system, the operation comprising:
identifying a first evidence related to a feature, wherein the question answering system has identified the feature as being relevant to answering a first type of question of a plurality of types of questions;
generating an updated response for the first question based on the changed first evidence responsive to: (i) detecting a change in the first evidence, and (ii) determining that a confidence score of a first response generated for a first question, of the first type of question, exceeds a confidence threshold; and
notifying a user of the change in the first evidence, wherein the updated response is generated upon approval of the user.
18. The non-transitory computer-readable storage medium of claim 17 , the operation further comprising monitoring the first evidence to detect the change.
19. The non-transitory computer-readable storage medium of claim 18 , wherein the first evidence is monitored upon determining that a relevance score of the first evidence exceeds a relevance threshold, wherein the first evidence comprises a content of at least one of: (i) a document, (ii) a variable, (iii) an article, and (iv) a web page, and wherein the first question is part of a first case comprising at least the first question.