Automatic semantic template generation method and device and semantic analysis method and system

Abstract

The invention discloses an automatic semantic template generation method and device and a semantic analysis method and system. The automatic semantic template generation method includes the steps of establishing a semantic slot library, wherein a plurality of semantic slots are stored in the semantic slot library, and each semantic slot comprises synonym sets and a semantic analysis result; establishing a rule base, wherein a plurality of rules are stored in the rule base, and each rule comprises at least one semantic slot; analyzing a rule in the rule base to obtain the semantic slots included in the rule; searching the semantic slot library for the included semantic slots; generating a semantic template by loading the synonym sets of the semantic slots included in the rule; generating the semantic analysis result of the semantic template by loading the semantic analysis results of the semantic slots included in the rule to the rule. By means of the automatic semantic template generation method, the semantic template can be automatically generated through a computer according to the rule, the writing time of the semantic template is saved, and the semantic cloud can be easy on the aspect of the expansion field.

Description

technical field [0001] The invention relates to the field of computer semantic recognition, in particular to a semantic template automatic generation method and device, and a semantic analysis method and system. Background technique [0002] In recent years, Internet technology based on voice interaction has gradually emerged. Currently, application developers can use "voice cloud" and "semantic cloud" to create an intelligent voice application. "Speech cloud" helps realize speech recognition and text conversion, while "semantic cloud" helps process and understand these texts, and provides semantic analysis functions for application developers. Specifically, the "Semantic Cloud" platform can provide application developers with a cloud-based semantic analysis management platform and a front-end SDK development kit. Developers can use this SDK to embed natural language understanding capabilities in their applications. Application developers can access or call the API to use ...

AI Technical Summary

Problems solved by technology

At present, semantic templates need to be manually written, which is time-consuming and labor-intensive. They are not very tolerant to redundancy and noise contained in the text, and have poor reusability, so that it is necessary to add attributes or add a new field on the basis of existing fields. When using a semantic template, it is necessary to manually write the semantic template repeatedly, which consumes a lot of time, making it difficult for the "semantic cloud" to expand the field

Images