Abstract: Due to the shortcomings of current semantic similarity metric not fully using the semantic information of ontology and its complex computation, this paper proposes a new semantic similarity metric calculation method based on the feature of concept, which makes full use of the information of ontology structure. It expresses each concept as a set of features according to the hierarchy of ontology, and introduces a width influencing factor as the coefficient of each feature. It obtains the semantic similarity through calculating the similarity between two sets. It introduces a depth influencing factor, and amends the semantic metric to a more understandable form. Theoretical analysis and experimental results show that the metric is simple, and the results are close to human judgment.

Key words: semantic Web, ontology, concept feature, feature attribute, semantic similarity

摘要： 现有语义相似度计算方法没有充分利用本体中的语义信息，且计算方法复杂。为此，提出一种基于概念特征的语义相似度计算方法。根据概念在本体中的层次结构来确定特征集合，引入宽度影响因子，给每个特征赋予不同的权值，通过计算2个概念特征集合间的相似度得到概念的相似度，引入深度影响因子，将相似度公式表示成更直观的形式。实验结果表明，该方法计算简便，且比较接近人类主观的判断值。

关键词: 语义Web, 本体, 概念特征, 特征属性, 语义相似度

CLC Number: 

• TP391.1