User-defined database interaction dialogue generation method and system

Abstract

The invention discloses a user-defined database interaction dialogue generation method and system, and the method comprises the steps of obtaining an action of a previous dialogue round, receiving a text inputted by a user if the action is a receiving operation action, and extracting an intention and an entity of the text inputted by the user through a natural language understanding module to update a dialogue state so as to generate the current dialogue information; converting the updated dialogue state into a binary vector, synchronously updating the binary vector into the historical dialogue information, and generating an execution action according to the current dialogue information and the historical dialogue information; if the execution action is not the receiving operation action,judging whether the execution action needs to retrieve the database or not; if yes, taking the user expectation as a retrieval condition to retrieve the database, and fusing a retrieval result with apreset template to generate a reply action for the user; if not, generating a reply action to the user directly by selecting any one of the predefined templates. According to the method, the user autonomously selects a database query result, so that the user experience is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method and a system for generating a dialog for user-defined database interaction. Background technique [0002] In recent years, with the vigorous development of the artificial intelligence industry, human-computer dialogue, as a sub-direction in the field of artificial intelligence, has attracted more and more attention. In layman's terms, human-computer dialogue technology is to allow human natural language to interact with computers in real time. The task-oriented human-computer dialogue technology can be applied to business scenarios such as booking hotels, meals, air tickets, and train tickets, and has high application value and commercial value. Traditional task-based multi-turn dialogue systems mainly include the following four modules: natural language understanding, dialogue state tracking, strategy formulation, and natural language generation. Among ...

AI Technical Summary

Problems solved by technology

But this method is rigid and inflexible. Users can only see a fixed template style, which cannot be well combined with the actual dialogue scene, so the user experience is not friendly enough.

Images