Cognitive question answering pipeline calibrating

A system comprising:

• a computer readable storage medium with program instructions stored thereon;
• and
• one or more processors configured to execute the program instructions to perform a method comprising:
• generating, based on a set of test questions and on a first answer key associated with the set of test questions, a first set of test answer data for a first answering pipeline;
• generating, for the first answering pipeline and based on the first set of test answer data, a first blending profile, the first blending profile comprising metadata about a confidence of the first answering pipeline;
• generating, based on the set of test questions and on a second answer key associated with the set of test questions, a second set of test answer data for a second answering pipeline;
• and
• generating, for the second answering pipeline and based on the second set of test answer data, a second blending profile, the second blending profile comprising metadata about a confidence of the second answering pipeline.

A computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a computer processing circuit to cause the circuit to perform the method comprising:

• generating, based on a set of test questions and on a first answer key associated with the set of test questions, a first set of test answer data for a first answering pipeline;
• generating, for the first answering pipeline and based on the first set of test answer data, a first blending profile, the first blending profile comprising metadata about a confidence of the first answering pipeline;
• generating, based on the set of test questions and on a second answer key associated with the set of test questions, a second set of test answer data for a second answering pipeline;
• and
• generating, for the second answering pipeline and based on the second set of test answer data, a second blending profile, the second blending profile comprising metadata about a confidence of the second answering pipeline.