Interactive labeling method and system for sentence pattern structure diagram analysis

Abstract

The invention provides interactive labeling method and system for sentence pattern structure diagram analysis. The method comprises the steps: carrying out the extension of a conventional regular expression, and obtaining a novel regular expression; constructing a sentence pattern rule set by using a novel regular expression; performing word segmentation and part-of-speech tagging on the to-be-tagged text to generate a word object sequence; and performing single-layer sentence pattern structure analysis on the word object sequence by using the sentence pattern rule set, and prompting an analysis result in a suspension box near the to-be-annotated text in a sentence pattern structure expression form. According to the interactive annotation method and system for sentence pattern structure diagram analysis, an interactive annotation environment is constructed, corpus annotation adopts a layer-by-layer analysis and layer-by-layer annotation mode, and under the condition that the analysis result of the system is correct, segmentation of all single-layer components can be completed through one-time mouse operation, so that the annotation efficiency is improved.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to an interactive labeling method and system for graphical analysis of sentence structure. Background technique [0002] In the research of theoretical linguistics and computational linguistics, treebank is a very important resource. In theoretical linguistics, treebanks provide linguists with a large amount of data based on real languages, laying a solid foundation for improving and verifying traditional linguistic theories. In computational linguistics, automatic syntax analysis and various upper-level applications depend on the scale and quality of the treebank. [0003] The mainstream treebank is mainly constructed based on the phrase structure grammar system and the dependency structure grammar system. After years of research, the accuracy rate of the automatic syntax analysis algorithm under the two grammar systems has been able to reach about...

AI Technical Summary

Problems solved by technology

However, there are two main problems in the purely manual labeling method: (1) The dichotomous labeling process does not conform to the sentence-based idea, but is closer to the idea of ​​​​phrase structure; (2) The sentence components need to be segmented one by one, and the labeling efficiency is too low

Images