Evaluating expert opinions in a question and answer system

1. An information handling system comprising:

one or more processors;

a memory coupled to at least one of the processors;

a set of instructions stored in the memory and executed by at least one of the processors, wherein the set of instructions perform actions of:

receiving a plurality of responses at a question and answering (QA) system, wherein each of the responses includes an evaluation of a passage selected from a plurality of passages corresponding to a question and an answer, and wherein each of the responses is received from one of a plurality of subject matter experts (SMEs);

calculating a plurality of passage evaluation scores, each of the plurality of passage evaluation scores based on the received responses, wherein each of the passage evaluation scores correspond to a different one of the plurality of passages;

associating the plurality of passage evaluation scores to the question and the answer, wherein each of the passage evaluation scores signifies a relevance of one of the plurality of passages to the question and the answer; and

training a model utilized by the QA system, wherein the training is based in part on the plurality of passage evaluation scores associated with the question and the answer.

2. The information handling system of claim 1 wherein the actions further comprise:

identifying a selected one of the plurality of responses that was received from a selected one of the plurality of SMEs;

evaluating the selected response by comparing the selected response to the plurality of responses; and

updating, in the memory, a SME confidence score corresponding to the selected SME based on the evaluation.

3. The information handling system of claim 2 wherein the selected SME is a new SME, and wherein the actions further comprise:

including the new SME in an SME cluster of a plurality of SMEs with a common area of expertise;

receiving one or more of the plurality of responses from the selected SME and the plurality of SMEs included in the SME cluster;

evaluating the selected SME's responses by comparing the selected SME's responses to the responses received from other SMEs included in the SME cluster; and

wherein the updating of the SME confidence score includes initializing the SME confidence score corresponding to the selected SME based on the evaluation.

4. The information handling system of claim 2 wherein the selected SME is a previously evaluated SME, and wherein the actions further comprise:

receiving one or more of the plurality of responses from the selected SME and a plurality of SMEs included in a SME cluster;

evaluating the selected SME's responses by comparing the selected SME's responses to the responses received from other SMEs included in the SME cluster; and

wherein the updating of the SME confidence score includes initializing the SME confidence score corresponding to the selected SME based on the evaluation.

5. A computer program product stored in a non-transitory computer readable storage medium, comprising computer instructions that, when executed by an information handling system, causes the information handling system to perform actions comprising:

receiving a plurality of responses at a question and answering (QA) system, wherein each of the responses includes an evaluation of a passage selected from a plurality of passages corresponding to a question and an answer, and where each of the responses is received from one of a plurality of subject matter experts (SMEs);

calculating a plurality of passage evaluation scores, each of the plurality of passage evaluation scores based on the received responses, wherein each of the passage evaluation scores correspond to a different one of the plurality of passages;

associating the plurality of passage evaluation scores to the question and the answer, wherein each of the passage evaluation scores signifies a relevance of one of the plurality of passages to the question and the answer; and

training a model utilized by the QA system, wherein the training is based in part on the plurality of passage evaluation scores associated with the question and the answer.

6. The computer program product of claim 5 wherein the actions further comprise:

identifying a selected one of the plurality of responses that was received from a selected one of the plurality of SMEs;

evaluating the selected response by comparing the selected response to the plurality of responses; and

updating, in the memory, a SME confidence score corresponding to the selected SME based on the evaluation.

7. The computer program product of claim 6 wherein the selected SME is a new SME, and wherein the actions further comprise:

including the new SME in an SME cluster of a plurality of SMEs with a common area of expertise;

receiving one or more of the plurality of responses from the selected SME and the plurality of SMEs included in the SME cluster;

evaluating the selected SME's responses by comparing the selected SME's responses to the responses received from other SMEs included in the SME cluster; and

wherein the updating of the SME confidence score includes initializing the SME confidence score corresponding to the selected SME based on the evaluation.

8. The computer program product of claim 6 wherein the selected SME is a previously evaluated SME, and wherein the actions further comprise:

receiving one or more of the plurality of responses from the selected SME and a plurality of SMEs included in a SME cluster;

evaluating the selected SME's responses by comparing the selected SME's responses to the responses received from other SMEs included in the SME cluster; and

wherein the updating of the SME confidence score includes initializing the SME confidence score corresponding to the selected SME based on the evaluation.