Cognitive computing techniques for optimal collaboration participant selection

A computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation comprising:

• receiving a domain for a collaborative session;
• retrieving a plurality of trait profiles corresponding to potential participants, wherein each of the plurality of trait profiles comprises a social characteristic of the respective potential participant and a determined dynamism, wherein the dynamism is based on a similarity of a subject matter of contributions made by the respective potential participant across collaborative sessions and the similarity of the subject matter of the contributions made by the respective potential participant to a subject matter of contributions from other potential participants in the collaborative sessions;
• determining, using one or more cognitive computing techniques, a subset of trait profiles from the plurality of trait profiles, based on the received domain and further based on comparing each respective trait profile with the rest of the plurality of trait profiles;
• providing the subset of trait profiles;
• receiving a rejection of a first trait profile in the subset of trait profiles;
• and
• refining the one or more cognitive computing techniques based on the rejection.

A method comprising:

• receiving a domain for a first collaborative session;
• retrieving a plurality of trait profiles associated with potential participants;
• determining, using a machine learning model, a subset of trait profiles from the plurality of trait profiles, based on the received domain and further based on comparing each respective trait profile with the rest of the plurality of trait profiles;
• providing the subset of trait profiles;
• receiving an indication that the subset of trait profiles is acceptable;
• receiving a request to initiate the first collaborative session with a plurality of participants corresponding to the subset of trait profiles;
• initializing a plurality of sensor devices;
• detecting, using at least one of the plurality of sensor devices, a first contribution from a first participant in the plurality of participants;
• classifying the first contribution by processing the first contribution using one or more cognitive computing techniques;
• determining a dynamism of the first participant based on:
• (i) a similarity of a subject matter of the first contribution to a subject matter of one or more previous contributions from the first participant in the first collaborative session and at least one previous collaborative session, and
• (ii) a similarity of the subject matter of the first contribution to a subject matter of one or more previous contributions from at least one other participant in the first collaborative session;
• identifying a trait profile in the subset of trait profiles corresponding to the first participant;
• and
• updating the identified trait profile based on the classified first contribution and the determined dynamism.