判别分析k最近邻判别分析法1
Discriminant analysis the K-Nearest Neighbors Discriminant analysis Method 1
投稿时间:2025-08-27  修订日期:2025-08-27
DOI:
中文关键词:  k最近邻判别分析  欧氏距离  多数投票  加权投票  判别规则
英文关键词:k-nearest neighbors (k-NN) discriminant analysis  Euclidean distance  majority voting  weighted voting  discriminant rule
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1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k最近邻判别分析法、确定k值与寻找k个最近邻、样品间的距离、多数投票或加权投票和基本思想;计算方法涉及三种距离的计算公式、判别准则、分类规则和特殊情况处理;两个实例中的资料分别是“三类鸢尾属植物在4项定量指标上的测定结果”与“五类作物在4项定量指标上的测定结果”;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? k-nearest neighbors (k-NN) discriminant analysis Method 1?using SAS software. The fundamental concepts?included k-NN discriminant analysis Method, determination of? k value?and identification of? k-nearest neighbors, distance metrics between samples, majority voting or weighted voting, and the core principles of the method. Computational methods?covered the formulas for three distance metrics, discriminant criteria, classification rules, and handling special cases. Two datasets?were analyzed: The first one was the measurements of?four quantitative traits?across?three Iris species, and the second one was the measurements of?four quantitative traits?across?five crop types. The SAS software was employed to perform k-NN discriminant analysis on both datasets, followed by detailed interpretation of the SAS output.
  查看/发表评论  下载PDF阅读器
关闭