最简多元最小上界算法研究

The Simplified Multielement Least Upper Bounds Algorithm Research

求概念的近似是近似查询中的关键问题,其质3决定了近似查询的质量。现有的基于概念最小上界和最大下界的近似查询方法,只考虑了异构本体概念间一对一的蕴涵关系,是无法得到概念的最佳近似,近似的质量有时是不可接受的。先把最小上近似转化为多元界中最小上界问题,再把最小上界转化为最简多元最小上界问题来研究。这种把求概念的最小上近似间接转化为求概念的最简多元最小上界方法,可以大大提高最小上近似的质量,进而提高近似查询的查准率和查全率。

For the approximate concept is similar to the query key issues, it's quality decision the approximate inquiries quality. Based on the existing concept of the smallest upper and lower bound of the largest approximate inquiries, only consider the concept of heterogeneous ontology contains between one-to-one relationship, it is impossible to approximate the best concept, Approximate sometimes the quality is unacceptable. First on the approximate minimum conversion to a pluralistic community, among the smallest on the border issue then the least upper bound into the smallest multi-SR on the question of research. This kind of concept for the smallest on the approximate indirect conversion for the sake of the most simple concept of pluralism on the smallest sector, can greatly enhance the smallest on the approximate quality and raise similar inquiries precision and recall rates.