论文部分内容阅读
提出采用混合多目标优化算法以解决优化三类不同的相似度度量技术(基于语法学的相似度度量,基于语言学的相似度度量和基于分类学的相似度度量)的映射结果集成的问题.比起传统的基于遗传算法的方法,本文提出的方法能够同时实现三个目标,即最大化映射的查全率recall、查准率precision和f-度量f-measure值,获取的本体映射结果能够避免对于查全率或是查准率的偏好.实验结果表明本文提出的方法是有效的.
A hybrid multi-objective optimization algorithm is proposed to solve the problem of integrating the mapping results of three different similarity measurement techniques (grammar-based similarity measure, linguistic similarity measure and taxonomy-based similarity measure). Compared with the traditional GA-based method, the proposed method can achieve three objectives simultaneously: maximizing recall recall, precision precision and f-measure of the mapping, and obtaining the ontology mapping result Avoid the preference for recall or accuracy.Experimental results show that the proposed method is effective.