Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (28): 110-112.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Ontology similarity computation based on spectral graph theory

LAN Meihui1,XU Jian1,SUN Yu2   

  1. 1.Department of Computer Science and Engineering,Qujing Normal University,Qujing,Yunnan 655011,China
    2.Department of Computer Science and Information Technology,Yunnan Normal University,Kunming 650092,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-01 Published:2011-10-01

基于谱图理论的本体相似度计算

兰美辉1,徐 坚1,孙 瑜2   

  1. 1.曲靖师范学院 计算机科学与工程学院,云南 曲靖 655011
    2.云南师范大学 计算机科学与信息技术学院,昆明 650092

Abstract: Research on semantic similarity of concepts is an important content in the fields of knowledge representation and information retrieval.It expresses information that is related to the ontology concept as a vector,and establishes an adjoin graph for the original ontology graph.It defines the edge set with ε-neighborhoods and the edge weights with the Gaussian kernel function.By computing the eigenvector that the second smallest eigenvalue of Laplacian matrix corresponds to,it can get ontology similarity calculation function.Experimental results show that the algorithm is effective.

Key words: ontology, concept similarity, spectral graph theory, Laplace Beltrami operator, Gaussian kernel function

摘要: 概念的语义相似度研究,是知识表示以及信息检索领域中的一个重要内容。将与某概念相关的信息表示为一个向量,建立原本体图的伴随图。用ε-领域方法定义边,用高斯核函数定义边的权值。通过计算图拉普拉斯矩阵的次小特征值对应的特征向量得到本体相似度计算函数。实验结果表明该算法是有效的。

关键词: 本体, 概念相似度, 谱图理论, 拉普拉斯Beltrami算子, 高斯核函数