首页|Construction of Preposition-verb Knowledge Base Based on Structure Retrieval

Construction of Preposition-verb Knowledge Base Based on Structure Retrieval

扫码查看
以往的介词知识库构建重视介词语义和介宾的搭配研究,鲜有对介动搭配进行系统研究及知识获取的工作。而汉语介词发达及动词是句子中心的特征决定了介动搭配研究的重要性。本研究基于结构检索技术,充分借助短语结构属性和结构信息,从大规模语料中抽取介动搭配16033对。并提出了介动搭配紧密度的度量方法,初步分析证明其远优于依靠绝对频次进行搭配度量的方法。
The construction of preposition knowledge base in the past had given more prominence to the study of semantics of prepositions as well as collocations of prepositions and their complements, while there had been few work on systematic research and knowledge extraction of preposition-verb collocations。 The characteristics of Chinese that prepositions are well-developed and that the heads of sentences being verbs determine that the study of preposition-verb collocations is crucial。 Being based on a technology of structures retrieving and fully utilizing phrasal structure attributes as well as information of structures, we have extracted 16033 pairs of preposition-verb collocations from a large-scale corpus, and came up with a method for measuring the closeness of a preposition-verb collocation, which was preliminarily proved to be superior to the method of absolute frequency。

Preposition-verb collocationStructure retrievalCloseness of preposition-verb collocation

Wang Chengwen、Rao Gaoqi、Xun Endong

展开 >

School of Information Science,Beijing Language and Culture University,Beijing, 100083

Research Institute of International Chinese Language Education, Beijing Language and Culture University,Beijing, 100083

Chinese National Conference on Computational Linguistic

Hohhot(CN)

20th Chinese National Conference on Computational Linguistic

513-524

2021