Knowledge Parsing System and Method Based on Dependency Syntax Tree

Abstract

The invention proposes a knowledge parsing system and method based on a dependency syntax tree. A knowledge parsing system based on a dependency syntax tree, including: a knowledge base module and a parsing module. The knowledge parsing method based on the dependency syntax tree proposed by the present invention enables the knowledge points in the Chinese context to be clearly defined for accurate parsing. The knowledge base can be dynamically maintained in real time, clear and controllable, and unreasonable problems can be directly located and solved, unlike traditional deep learning models that cannot be analyzed like "black boxes". The knowledge analysis scene is no longer limited by the diversity and complexity of Chinese grammar / syntax. As long as the resources of the knowledge base are sufficiently high-quality and comprehensive, the scene application requirements can be met to the greatest extent.

Description

technical field [0001] The invention relates to the field of natural language processing, in particular to a knowledge parsing system and method based on a dependency syntax tree. Background technique [0002] Dependency parsing is an important part of natural language processing. Dependency syntax can reflect the internal logic of natural language, and it is a syntax theory that breaks through the constraints of languages ​​and exists in every language family. The concept of "dependency syntax" was first proposed by the Indian linguist Panini in the 4th century BC. The original intention was to study grammar, syntax, semantics and dependency patterns in different categories. It was published by French linguist Lucien Tesniere in 1959 The book "Basics of Structural Syntax" has always been considered as the theoretical basis of modern dependency syntax. In 1970, Robinson proposed four dependency axioms based on dependency syntax, which laid the theoretical foundation for dep...

AI Technical Summary

Problems solved by technology

However, there are some problems in the existing corpus. For example, the overall development is unbalanced, which is mainly manifested in the large difference in the number of written and spoken corpora, because the collection and sampling process of spoken corpus is relatively complicated and cumbersome.

The accuracy of the corpus cannot be guaranteed. The huge corpus contains many sentences that need to be modified. The root cause is the lack of effective self-examination methods

Images