基于规则的中文组块内部关系分析  

Rule-based Analysis of the Internal Relations of Chinese Chunks

作　　者：王贵荣 荀恩东[1] 饶高琦[2] WANG Guirong;XUN Endong;RAO Gaoqi(School of Information Science, Beijing Language and Culture University, Beijing 100083, China;Research Institute of International Chinese Language Education, Beijing Language and Culture University, Beijing 100083, China)

机构地区：[1]北京语言大学信息科学学院,北京100083 [2]北京语言大学汉语国际教育研究院,北京100083

出　　处：《郑州大学学报（理学版）》2022年第3期28-33,共6页Journal of Zhengzhou University:Natural Science Edition

基　　金：国家自然科学基金项目(62076038)。

摘　　要：组块分析是浅层句法分析的典型任务,但目前的研究大多集中于组块边界识别和组块句法功能研究,缺少对组块内部关系的分析。采用基于规则的多结构融合方法进行块内关系分析,即将规则集与有限状态自动机、树结构、网格结构以及搭配知识进行有效融合,分析结果以三元组的形式表示。以宾州中文树库中gold standard数据集的共计2005句作为测试集进行了实验,最终F1值达到了85.82%。所提方法加深了句法分析的深度,推动了传统组块分析向完全句法分析的发展。Chunk parsing was a typical task of shallow syntactic analysis,but most of the current studies focused on the recognition of chunk boundary and chunk syntactic functions,and lacked the analysis of the internal relationship of chunks.A rule-based multi-structure fusion method was used to analyze the syntactic relationship between words in the chunk.The rule set was effectively fused with the FSA,tree,lattice and collocation knowledge,and the analysis results were expressed in the form of triples.The gold standard data were selected from the CTB 8.0 version as the test set,a total of 2005 sentences,the F1 value reached 85.82%.This method deepened the depth of syntactic parsing,and promoted the development of chunk parsing to complete syntactic parsing.

关 键 词：组块 规则 块内分析 语言资源 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]