System and method for identifying themes in interactive communications

A computer-implemented method for processing an interactive communication, the method comprising:

• inferring, from the interactive communication and based on a first trained machine learning model, a topic associated with the interactive communication;
• extracting, from the interactive communication, keywords associated with the topic;
• ranking the keywords, based on a frequency of occurrence within the interactive communication;
• selecting the N highest ranked keywords from the interactive communication;
• comparing the N highest ranked keywords to a database of previously extracted N highest ranked keywords, occurring over a given time period, to identify new keywords;
• labeling, based on the comparison, the topic as an emerging topic that has been articulated in the interactive communication;
• extracting, based on a second trained machine learning model, similar previous customers associated with the emerging topic;
• and
• suggesting, based on previously successful interactive communication outcomes with the emerging topic for the similar previous customers, a list of actions to resolve issues associated with the emerging topic to a call center call agent for use in a dialog of the interactive communication.

A non-transitory computer-readable device having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:

• inferring, from an interactive communication and based on a first trained machine learning model, a topic associated with the interactive communication;
• extracting, from the interactive communication, keywords associated with the topic;
• ranking the keywords, based on a frequency of occurrence within the interactive communication;
• selecting the N highest ranked keywords from the interactive communication;
• comparing the N highest ranked keywords to a database of previously extracted N highest ranked keywords, occurring over a given time period, to identify new keywords;
• labeling, based on the comparison, the topic as an emerging topic that has been articulated in the interactive communication;
• extracting, based on a second trained machine learning model, similar previous customers associated with the emerging topic;
• and
• suggesting, based on previously successful interactive communication outcomes with the emerging topic for the similar previous customers, a list of actions to resolve issues associated with the emerging topic to a call center call agent for use in a dialog of the interactive communication.