Dialogue management method and a related device

Abstract

The invention discloses a dialogue management method, which comprises the following steps: carrying out natural language understanding processing on a received natural language text to obtain intention analysis data; searching a corresponding intention root node in a node logic tree according to the intention analysis data, and performing slot filling processing on an intention slot of the intention root node according to word recognition data to obtain preliminary intention analysis data; wherein the node logic tree is obtained by performing configuration processing according to configurationinformation; judging whether the preliminary intention analysis data is complete intention analysis data or not according to the node logic tree; if yes, according to the complete intention analysisdata, outputting answer data according to a preset path. Through the node logic tree obtained through configuration, dialogue management is achieved, the logicality in dialogue management is improved,and the online speed of dialogue management is increased. The invention further discloses a dialogue management system, dialogue management equipment and a computer readable storage medium which havethe above beneficial effects.

Description

technical field [0001] The present application relates to the technical field of natural language processing, and in particular to a dialog management method, a dialog management system, a dialog management device, and a computer-readable storage medium. Background technique [0002] With the continuous development of information technology, man-machine dialogue technology is still the key development technology in artificial intelligence. Among them, the multi-round dialogue technology is one of the core technologies for realizing the intelligent human-computer interaction system. It has a wide range of application scenarios and can be directly used in specific business processes, such as hotel reservations, train ticket reservations, etc. [0003] At present, the existing technology provides a variety of technologies to drive multiple rounds of dialogue, including driving based on knowledge base search and driving based on deep learning models. However, there are obvious ...

AI Technical Summary

Problems solved by technology

However, there are obvious deficiencies in the existing technology in how to efficiently, concisely and logically combine with specific services to achieve multiple rounds of dialogue management

The method based on knowledge base search can realize a wide range of commonly used sentences to answer questions, but it cannot be well solved on logical business problems

However, the method based on the deep learning model will seriously affect the accuracy of multiple rounds of dialogue when the amount of training data is insufficient, and has extremely high requirements for data labeling, which is not conducive to rapid application in specific business fields, and the early construction efficiency is low.

Images