Method for verb hierarchical classification of multi-verb Chinese concept compound chunk

Abstract

The present invention relates to a method for verb hierarchical classification of a multi-verb Chinese concept compound chunk. The method comprises the following steps of (1) performing word segmentation and part-of-speech tagging on a Chinese sentence; (2) dividing the sentence into minor sentences; (3) performing hierarchical division on multiple verbs of the minor sentences; (4) dividing the verbs into two types; (5) analyzing the verbs by utilization of an improved CCC analyzer; and (6) analyzing the multi-verb minor sentences until the queue element is empty. According to the method, sentence overall hierarchical error caused by verb hierarchical analysis error can be reduced, so that accuracy of chunk analysis can be improved, a good foundation is provided for subsequent syntax and semantics analysis of complex sentences, and requirements of practical application can be well satisfied.

Description

technical field [0001] The invention belongs to the technical field of Chinese information processing, and in particular relates to a method for classifying verb levels of multi-verb Chinese concept compound blocks. Background technique [0002] Syntactic analysis is a key issue in the deep processing of natural language. At present, the complete syntactic analysis of Chinese sentences is still in its infancy, and the chunking analysis proposed as an intermediate step of complete syntactic analysis has become the focus of research at this stage. [0003] The study of Chinese chunks has gone through a gradual development process, initially focusing on the study of basic nouns, prepositional phrases and so on. With the development of corpus, researchers have proposed different block description systems. The analysis and research at the present stage only focus on the relatively simple and functionally important components in the sentence. As an intermediate step between lexic...

AI Technical Summary

Problems solved by technology

[0004] The concept compound block is one of many block description systems. Based on the research and analysis of it, it is found that when the block analyzer for the concept compound block at the present stage analyzes and processes sentences containing multiple verbs, The errors mainly exist in the inaccurate recognition of the boundaries between the various components. For example, the words originally belonging to the subject part are divided into the predicate or adverbial components, and the content originally belonging to the object is divided into the predicate part. Hierarchy analysis error

Images