Semantic fusion method and system based on micro-service, and medium

Abstract

The invention provides a semantic fusion method and system based on a micro-service, and a medium, and relates to the technical field of man-machine interaction. The method comprises the following steps: S1, constructing a plurality of business skill semantic understanding micro-services according to business demands; S2, taking a user corpus text as input of a plurality of business skill semantic understanding micro-services to obtain a plurality of semantic understanding results; and S3, conducting semantic decision making on the multiple semantic understanding results through confidence levels, and acquiring an optimal semantic result at the same time. According to the method, a plurality of semantic understanding function modules can be fused by adopting a tree structure, the problems that a large memory is occupied when a plurality of business skill semantic understanding models are loaded, different versions of the same installation package are incompatible and service engineering is not facilitated are solved, dialogue performance is improved, and meanwhile, good user experience of man-machine dialogue is guaranteed.

Description

technical field [0001] The present invention relates to the technical field of human-computer interaction, in particular to a semantic fusion method, system and medium based on microservices. Background technique [0002] Natural language is the crystallization of human wisdom, and it has been a long-standing expectation for people to interact with computers in the form of natural language. With the development of natural language processing technology and the rise of deep learning methods, the human-computer dialogue system has become a new research hotspot. The fusion of semantic understanding is a very important part of human-computer dialogue. If the semantic understanding fusion method is not good, it may affect the accuracy and performance of the dialogue, and even fail to guarantee a good user experience of human-computer dialogue. [0003] The invention patent with the publication number CN111665941A discloses a virtual experiment-oriented multi-modal semantic fusio...

AI Technical Summary

Problems solved by technology

[0004] There are the following deficiencies in the integration of semantic understanding in the existing technology: the existing technology is mainly in the form of code, and there are problems that the loading of multiple business skill semantic understanding models takes up a lot of memory, and different versions of the same installation package are not compatible and are not conducive to service engineering;

Images