判别分析k最近邻判别分析法4
Discriminant analysis the K-Nearest Neighbors Discriminant Analysis Method 4
投稿时间:2025-08-27  修订日期:2025-08-27
DOI:
中文关键词:  距离度量  k值选择  特征标准化  k最近邻判别分析  懒惰学习
英文关键词:Distance metrics  K-value selection  Feature standardization  K-nearest neighbor discriminant analysis  Lazy learning
基金项目:
作者单位地址
胡良平* 军事科学院研究生院 E-mail:lphu927@163.com
摘要点击次数:
全文下载次数:
中文摘要:
      本文目的是介绍与k最近邻判别分析法4有关的基本概念、计算方法、两个实例及其用SAS实现计算的方法。基本概念包括距离度量、k值选择、特征标准化、分类与回归、懒惰学习;计算方法涉及基本定义和计算公式;两个实例中的资料分别是“三个层段的样品5个定量指标的测定结果”和“美国三大制造商生产的早餐麦片相关数据”;借助SAS软件,对两个实例中的数据进行了k最近邻判别分析,对SAS输出结果给出了解释。
英文摘要:
      The purpose of this paper was to introduce the fundamental concepts, computational methods, two practical examples, and the implementation of calculations using SAS software related to the k-nearest neighbor (k-NN) discriminant analysis method 4. The fundamental concepts included distance metrics, selection of k-value, feature standardization, classification and regression, and lazy learning. The computational methods covered basic definitions and relevant formulas. The two datasets used in the examples were "measurement results of five quantitative indicators from samples in three stratigraphic sections" and "data related to breakfast cereals produced by three major U.S. manufacturers." Using SAS software, k-nearest neighbor discriminant analysis was performed on the data from these two examples, and the SAS output results were interpreted.
  查看/发表评论  下载PDF阅读器
关闭