| 摘 要: 对于一条给定的句子，目标词识别就是识别出句子中能够激起语义场景的目标词。针对目标词识别中的特
征选择问题，本文把目标词识别任务看作是一个分类问题，在传统的词法特征和上下文特征基础上，加入了句法结构特
征设计特征模板，识别句子中的目标词。在汉语框架网的标注语料集上进行测试，实验结果表明，相比于传统的词特
征，基于句法结构分析的目标词识别率有显著地提升。 |
| 关键词: 框架语义分析；目标词识别；句法结构分析；汉语框架网 |
|
中图分类号: TP391
文献标识码: A
|
|
| A Study on Target Word Identification Based on Syntactic Structure Analysis |
|
ZHANG Lifan
|
( School of Computer Engineer, Taiyuan Institute of Technology, Taiyuan 030008, China)
|
| Abstract: For a given sentence,target words identification is to identify a target word that can evoke the semantic scene.
Aiming at the feature selection problem in target words identification,it is considered as a classification problem.On the basis
of traditional lexical features and contextual features,the syntactic structure feature template is added to identify the target
words in the sentence.Compared to the traditional word features,the results show that the rate of target words identification
based on syntactic structure analysis is significantly improved. |
| Keywords: frame-semantic analysis;target word identification;syntactic structure analysis;Chinese FrameNet |
