双极隶属度F_KNN分类算法在算法管理中的应用

APPLYING TWO-BEST-MEMBERSHIP F_KNN CLASSIFICATION ALGORITHM IN ALGORITHM MANAGEMENT

算法管理是信息融合领域中一个新的研究课题,其过程涉及到对信息融合数据源的分类问题。针对信息融合数据源样本的特征,提出一种双极隶属度F_KNN分类算法,通过邻居质量优先隶属度函数和双极隶属度判决规则,得到理想的数据源分类结果。

Algorithm management is a new aspect in information fusion area, its process involves the classification of data sources of infor- mation fusion. According to the characteristic of data sources sample, a Two-Best-Membership F KNN classification algorithm is proposed. Neighbours-Quality-First membership function and Two-Best-Membership judging rules are used to get ideal classification results of data sources.