Semantic relationship recognition method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a semantic relationship recognition method and device, electronic equipment and a readable storage medium, and relates to the technical field of big data and cloud. The method comprises the steps of: obtaining a to-be-processed text; performing context analysis on the to-be-processed text to obtain a grammatical relationship sequence contained in the to-be-processed text; matching the grammatical relationship sequence with a class sequence rule base of a specific semantic relationship, and determining whether the to-be-processed text contains the specific semantic relationship or not based on a matching result, wherein the class sequence rule base comprises a grammatical relationship sequence rule corresponding to the specific semantic relationship, and the grammatical relationship sequence rule is obtained by performing class sequence rule mining of the grammatical relationship on a plurality of sample texts containing the specific semantic relationship. In the embodiment of the invention, the grammatical relationship sequence rule is obtained by mining the class sequence rule for performing the grammatical relationship, so that the method does not depend on the rule and a formulating mode, the generalization ability of the model to recognize the semantic relationship is improved, and the method has good operability.

Description

technical field [0001] The present application relates to the technical field of big data and cloud technology. Specifically, the present application relates to a semantic relationship recognition method, device, electronic equipment and readable storage medium. Background technique [0002] With the development of language and the continuous change of Internet terminology, specific semantic relationship recognition is an important task for natural language understanding, including knowledge graph construction, domain knowledge base construction, relationship chain recognition, automatic construction of entry labels, etc. The application scenarios are inseparable from the identification of specific semantic relations. Specific semantic relations include upper and lower meaning relations, total score relations, synonymous relations, antisense relations, etc., as well as species relations, composition relations, etc. defined from other logical relations, covering a wide range ...

AI Technical Summary

Problems solved by technology

However, it is found that the semantic relationship recognition method based on pattern matching can only extract the explicit relationship with artificially specified rules, and many implicit patterns of semantic relationship are flexible. This method greatly reduces the coverage of relationship extraction, and at the same time Rule expansion relies on experts and prior knowledge bases, which increases labor costs and time consumption, and online updates are not timely enough; while machine learning-based methods mainly rely on feature engineering, which often requires cumbersome feature selection and feature extraction. Computationally expensive and prone to error propagation issues

Images