• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2021, Vol. 43 ›› Issue (07): 1316-1323.

• 人工智能与数据挖掘 • 上一篇    下一篇

基于二元与三元模型相结合的句法规则层次化分析算法

张海玲,邵玉斌,贾继康,龙华,杜庆治   

  1. (昆明理工大学信息工程与自动化学院,云南 昆明 650500)
  • 收稿日期:2020-02-24 修回日期:2020-07-21 接受日期:2021-07-25 出版日期:2021-07-25 发布日期:2021-08-17
  • 基金资助:
    国家自然科学基金(61761025)

A hierarchical analysis algorithm of syntactic rules based on binary and ternary word models

ZHANG Hai-ling,SHAO Yu-bin,JIA Ji-kang,LONG Hua,DU Qing-zhi   

  1. (Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)



  • Received:2020-02-24 Revised:2020-07-21 Accepted:2021-07-25 Online:2021-07-25 Published:2021-08-17

摘要: 句法分析是自然语言处理领域中应用前景非常广阔的一个研究方向。针对目前句法分析多数是从字、词的角度出发且存在诸多不足,提出了二、三元词模型相结合的句法规则层次化分析算法,并结合分词、词性标注以及句子组织信息之间的结合度来解决词元间优先合成的问题,同时利用句子成分之间的语法结构关系对词性、词序的影响,实现句法规则的层次化分析实验。实验结果表明,二元与三元词模型相结合的句法规则层次化分析算法相比于独立二、三元词模型,准确率和召回率分别提高了82.04%和8083%,与现有基于二分结构句法分析的RNN-INT算法和词汇化模型算法相比,准确率和召回率均有明显提升。

关键词: 二、三元词, 句法结构, 词元结合模型, 层次化分析算法

Abstract: Syntactic analysis is a research direction with very broad application prospects in the field of natural language processing. In view of the fact that current syntactic analysis is mostly from the perspective of characters and words and there are many shortcomings, a hierarchical analysis algorithm of syntactic rules combining binary and ternary word models is proposed, and word segmentation, part-of-speech tagging and sentence organization information are combined to solve the problem of preferential synthesis between words. At the same time, the effect of grammatical structure relationship between sentence components on part-of-speech and word order is used to achieve a hierarchical analysis experiment of syntactic rules. The results show that the hierarchical rule analysis algorithm combining the binary and ternary word models is more accurate than the independent binary and ternary word models, and it improves the accuracy rate and recall rate to   and  , respectively. Compared with the existing   algorithm and lexical model algorithm based on binary structure syntax analysis, the accuracy and recall are significantly improved. 



Key words: binary and ternary word, syntactic structure, word element combination model, hierarchical analysis algorithm