Coordinated emotion representation and expression in multi-agent digital assistants

A computer program product comprising:

• one or more computer readable storage media;
• program instructions, stored on at least one of the one or more computer readable storage media, to monitor a conversation of a first user;
• program instructions, stored on at least one of the one or more computer readable storage media, to aggregate a dataset of human-agent interactions associated with the first user, wherein the dataset comprises text, audio, and video recordings associated with the first user and the human-agent interactions, where emotions of the first user are explicitly expressed or inferred;
• program instructions, stored on at least one of the one or more computer readable storage media, to determine an intervention opportunity is present in the conversation based on the interactions associated with the first user;
• responsive to determining the intervention opportunity is present, program instructions, stored on at least one of the one or more computer readable storage media, to retrieve a profile associated with the first user;
• program instructions, stored on at least one of the one or more computer readable storage media, to determine a context of the conversation based on the aggregation of the dataset of the human-agent interactions using sentiment analysis and named entity recognition;
• and
• based on the profile associated with the first user and the context of the conversation, program instructions, stored on at least one of the one or more computer readable storage media, to generate one or more emotion-based responses by utilizing affect-aware sentiment analysis to identify an emotional tone of the first user so the response reflects the emotions of the first user.

A computer system comprising:

• one or more computer processors;
• one or more computer readable memories;
• and
• one or more computer readable storage media;
• program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to monitor a conversation of a first user;
• program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to aggregate a dataset of human-agent interactions associated with the first user, wherein the dataset comprises text, audio, and video recordings associated with the first user and the human-agent interactions, where emotions of the first user are explicitly expressed or inferred;
• program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to determine an intervention opportunity is present in the conversation based on the interactions associated with the first user;
• responsive to determining the intervention opportunity is present, program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to retrieve a profile associated with the first user;
• program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to determine a context of the conversation based on the aggregation of the dataset of the human-agent interactions using sentiment analysis and named entity recognition;
• and
• based on the profile associated with the first user and the context of the conversation, program instructions, stored on at least one of the one or more computer readable storage media for execution by at least one of the one or more computer processors via at least one of the one or more memories, to generate one or more emotion-based responses by utilizing affect-aware sentiment analysis to identify an emotional tone of the first user so the response reflects the emotions of the first user.