System and method for using a question and answer engine

A system for using a question and answer engine to automatically generate responses to an end-user, the system comprising:

• a processor;
• machine-readable media including instructions which, when executed by the processor, cause the processor to:
• classify each utterance stored in a historical utterance database under at least one topic and at least one subtopic;
• feed the at least one topic and at least one subtopic for each utterance to a predictive clustering model;
• classify, via the predictive clustering model, each of a plurality of questions and corresponding answers stored in a question and answer database under at least one topic and at least one subtopic;
• feed the at least one topic and at least one subtopic for each of the plurality of questions and answers to a recommender module;
• receive a first user utterance and determining, via the recommender module, that the first user utterance is associated with a first topic and a first subtopic;
• identify a list of related questions stored in the question and answer database that are also classified under the first topic and the first subtopic, the list of related questions including a first question, the first question having a corresponding first answer;
• identify, from the related questions, the first question as being similar to the first user utterance by applying a similarity scorer to the first user utterance;
• and
• automatically generate a first response to the first user utterance based on the first answer.

A non-transitory computer-readable medium storing software comprising instructions executable by one or more processors which, upon such execution, cause the one or more processors to:

• classify each utterance stored in a historical utterance database under at least one topic and at least one subtopic;
• feed the at least one topic and at least one subtopic for each utterance to a predictive clustering model;
• classify, via the predictive clustering model, each of a plurality of questions and corresponding answers stored in a question and answer database under at least one topic and at least one subtopic;
• feed the at least one topic and at least one subtopic for each of the plurality of questions and answers to a recommender module;
• receive a first user utterance and determining, via the recommender module, that the first user utterance is associated with a first topic and a first subtopic;
• identify a list of related questions stored in the question and answer database that are also classified under the first topic and the first subtopic, the list of related questions including a first question, the first question having a corresponding first answer;
• identify, from the related questions, the first question as being similar to the first user utterance by applying a similarity scorer to the first user utterance;
• and
• automatically generate a first response to the first user utterance based on the first answer.