基于知识图的汉语词汇语义相似度计算.pdfVIP

23 3 Vol. 23, No. 3 2009 5 JOU RNAL OF CHINESE INFORMAT ION PROCESSIN G May, 2009 : 1003 0077( 2009) 03 0116 05 1 1 2 , , ( 1. 450011; 2. 450044) : 提出了一种基于知识图的汉语词汇相似度计算方法, 该方法以5知网6 2005 版为语义知识资源, 以知识图 为知识表 方法,在构造词图的基础上, 以知网中的语义关系为依据对词汇概念中的义原进行分类, 通过计算不同 类型义原的相似度得到概念的相似度; 为了对词汇相似度计算方法进行客观评价, 设计了词汇相似度计算方法的 量化评价模型; 采用该模型对所提出的计算方法进行评价, 试验结果证明此方法的有效度为 89. 1% : 计算机应用; 中文信息处理; 知识图; 知网; 语义相似度 : T P391 : A A New Measure of Semantic Similarity between Chinese Words Based on Knowledge Graphs 1 1 2 ZHANG Ruixia , ZHU Guiliang , YANG Guozeng ( 1. Department of Information Engineering, North China University of Water Conservancy and Elect ric Power, Zhengzhou, H enan 450011, China; 2. Department of Mathematics Zhengzhou teacher. s College, Zhengzhou, H enan 450044, China) Abstract: A new measure of semantic similarity betw een Chinese w ords is put forward on the basis of Knowledge Graphs. With H ow N et (2005) as the semantic knowledge resource and the Know ledge Graphs as know ledge repre sentation method, this method classifies / sememes0 based on their semantic roles in H ow N et, and then measures semantics sim ilarity among the co const ructed w ord graphs by the similarity of different kinds of/ sememe0 . In order to evaluate the proposed measures of semantic similarity, a new model is designed for quant itative evaluation reuslt. With the help of this evaluation model, the experiment result proves that the effective degree of the our semant ic measure is 89. 1% . Key words: computer application;