ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2019, Vol. 56 ›› Issue (12): 2600-2611.doi: 10.7544/issn1000-1239.2019.20180572

• 人工智能 • 上一篇    下一篇

粒向量与K近邻粒分类器

陈玉明,李伟   

  1. (厦门理工学院计算机与信息工程学院 福建厦门 361024) (cym0620@163.com)
  • 出版日期: 2019-12-01
  • 基金资助: 
    国家自然科学基金项目(61573297,61976183);福建省出国留学奖学基金项目;福建省自然科学基金项目(2016J01198,2019J01850);福建省教育厅A类项目(JA15363);厦门市科技计划指导项目(3502Z20179038)

Granular Vectors and K Nearest Neighbor Granular Classifiers

Chen Yuming, Li Wei   

  1. (College of Computer and Information Engineering, Xiamen University of Technology, Xiamen, Fujian 361024)
  • Online: 2019-12-01

摘要: K近邻(K nearest neighbor, KNN)分类器是一种经典的分类器,它简单而又有效,已经在人工智能与机器学习领域得到了广泛的应用.针对传统分类器难以处理不确定性数据的问题,研究样本单特征邻域粒化技术,构造粒的向量形式,提出一种基于粒向量的K近邻分类方法.该方法引入邻域粗糙集模型,对分类系统中的样本进行单特征邻域粒化,形成特征邻域粒子.并由多个特征邻域粒子构成一个粒向量,定义了多种粒向量运算算子,提出了2种粒向量距离:相对粒距离与绝对粒距离,证明了粒向量距离的单调性原理.进一步,基于粒向量距离定义了K近邻粒向量概念,提出了K近邻粒分类器.最后,结合UCI数据集,采用K近邻粒分类器与经典K近邻分类器进行比较测试.理论分析和实验表明:针对合适的粒化参数与k值,K近邻粒分类器具有较好的分类性能.

关键词: KNN分类器, 粒计算, 粒向量, 粒距离, 粒分类器

Abstract: K nearest neighbor (KNN) classifier is a classical, simple and effective classifier. It has been widely employed in the fields of artificial intelligence and machine learning. Aiming at the problem that traditional classifiers are difficult to deal with uncertain data, we study a technique of neighborhood granulation of samples on each atom feature, construct some granular vectors, and propose a K nearest neighbor classification method based on these granular vectors in this paper. The method introduces a neighborhood rough set model to granulate samples in a classification system, and the raw data can be converted into some feature neighborhood granules. Then, a granular vector is induced by a set of neighborhood granules, and several operators of granular vectors are defined. We present two metrics of granular vectors which are relative granular distance and absolute granular distance, respectively. The monotonicity of distance of granular vectors is proved. Furthermore, the concept of K nearest neighbor granular vector is defined based on the distance of granular vectors, and K nearest neighbor granular classifier is designed. Finally, the K nearest neighbor granular classifier is compared with the classical K nearest neighbor classifier using several UCI datasets. Theoretical analysis and experimental results show that the K nearest neighbor granular classifier has better classification performance under suitable granulation parameters and k values.

Key words: K nearest neighbor (KNN) classifier, granular computing, granular vector, granular distance, granular classifier

中图分类号: