摘要
通过最小化扩展的AUC量度得到最优排序函数,从而将本体图(或多本体结构图)中每个顶点映射成一个实数,通过比较本体顶点对应实数间的差值判断概念间的相似程度。实验表明:新算法对于本体相似度计算和在不同本体间建立映射是有效的。
By minimum extension of AUC metric, we obtained the optimal ranking function that mapped the vertices of ontological graphs （or multi-ontological graphs） into real numbers. Then we calculated the relative similarities of concepts by finding the differences between their corresponding values. The results show that the method is efficient for calculating ontological similarity gies. and building ontological mapping among different ontologies.
出处
《苏州科技学院学报（自然科学版）》
CAS
2013年第1期71-74,共4页
Journal of Suzhou University of Science and Technology （Natural Science Edition）
基金
国家自然科学基金资助项目(60903131)
