AI METHOD AND COMPUTER READABLE RECORDING MEDIUM FOR PROVIDING ANSWERS BASED ON HYBRID HIERARCHICAL CONVERSATION FLOW MODEL WITH CONVERSATION MANAGEMENT MODEL USING MACHINE LEARNING

摘要

Conversation Understanding A method is provided for providing answers based on a hybrid hierarchical conversation flow model that applies machine learning to conversation management technology performed by an AI service server. The method of the present disclosure includes generating a hybrid hierarchical dialogue flow model; Receiving a user's natural language input from a user terminal; Determining an intent of the user corresponding to the received natural language input; Determining a current intent group on the hybrid hierarchical conversation flow model; Calculating a matching rate of the intent of the user for each of the current intent group and a lower intent group of the current intent group; Determining whether the highest matching rate exceeds a predetermined threshold; When the highest matching rate exceeds a predetermined threshold, selecting an answer to the user input from the current intent group having the highest matching rate; And when the highest matching rate is below a predetermined threshold, switching to a deep learning mode and selecting an answer for a user input.