基于领域本体的文献模糊相似度算法研究  

Research on Fuzzy Similarity Algorithm Based on Domain Ontology

作　　者：宋雯斐[1] 王洋[2] 双林平[1] 

机构地区：[1]浙江工业职业技术学院,绍兴312000 [2]黑龙江科技学院,哈尔滨150027

出　　处：《图书情报工作》2011年第1期110-113,共4页Library and Information Service

摘　　要：利用分类主题一体化的主题词表构建领域本体,并通过概念间的关系定义及语义相似度公式,引入调整因子,确定概念相似度算法,再通过余弦系数法进一步得到文献间的相似度。对于本算法的结果,与领域专家预测的相似度进行比较,结果证实该算法有效。The paper builds a domain ontology by subject thesaurus integrated classified. Based on the ontology, the concept relation definition and semantic similarity formula, and combined with a adjustment factor, it computes the concept similarity. At last, the document similarity algorithm is defined by Cosine Coefficient. The results of document similarity computed by the algorithm are close to the similarities estimated by domain experts, so it is effective.

关 键 词：领域本体 文献相似度 概念相似度 

分 类 号：G202[文化科学—传播学]