基于多词块的框架元素语义核心词自动识别研究.PDFVIP

24 1 Vol. 24, No . 1 2010 1 JOU RNAL OF CHINESE INFORMAT ION PROCESSIN G Jan., 2010 : 1003-00 ( 2010) 01-0030-0 , , , ( , 030006) : 抽取 一个句子的核心依存图是对句子进行语义理解的有效途径在CFN 自动标注的基础 , 只能得到框 架依存图, 为了把框架依存图转换成框架核心依存图需要提取每 个框架元素的语义核心词该文提出了基于多词 块标注的框架元素语义核心词识别和提取方法, 通过对比分析, 给出了多词块和框架元素的融合策略, 并建立了在 多词块标注基础 提取框架元素语义核心词的规则集在6 1 个框架元素 的实验结果显示, 采用该文的方法 和规则集提取框架元素核心词的平均准确率和覆盖率分别为95. 58% 和82. 91% : 计算机应用; 中文信息处理; 框架元素; 语义核心词; 多词块 : T P391 : A Multi-Word Chunking Based Automatic Identification of the Semantic Core Word of the Frame Element LI Shuanghong, LI Ru, ZHONG Lijun, GUO Weiyu ( School of Computer Information Technology , Shanxi Universit y, T aiyuan 030006, China) Abstract: It is an effect ive w ay to understand the semantic information of a sent ence by extracting the frame kernel dependency graph from the sentence. It is necessary to ext ract semantic core w ords for each frame element to furt her establish the frame kernel dependency graph since we can only extract the f rame dependency graph from a sentence based on t he automatic annotation of CFN, This paper proposes a method to identify and extract the core w ords of frame elements by mult-i w ord chunk. On the basis of comparative analyzing results, we propose the strategy of integrating the multi-word chunk and f rame element and the rules to extract the core words of frame elements from the mult-i w ord chunk labeling . The experimental results from 6 1 frame elements show that the average precision and average coverage are 95. 58% and 82. 91% , respect ively. Key ords: computer application; Chinese information processing; Frame element; Semantic core words; Multi-word chunk