一种基于均值的多维样本空间分类器的设计与实现

分类算法是机器学习的一个重点,是有监督的学习.论文设计并实现了一种基于均值的多维样本空间分类器.对每个训练样本属性,计算该属性的类别X平均值和非类别X平均值的中值,作为类区分值,构建分类器;并进行分类测试,小于分离值的计为类别X,大于分离值的计为非类别X.然后根据类别X和非类别X属性的个数,由多数方决定最后的类别预测;...

Design and Implementation of a Classifier Based on Mean in Multidimensional Sample Space

Classification algorithm is a focus of machine learning,it's supervised learning.This paper designs and implements a classifier of multivariate sample space based on the mean.For each training sample attribute,the median value of the category X average value and the non-category X average value of the attribute are calculated,as the class distinction value,the classifier is constructed and the classification test is performed,the value smaller than the separation value is counted as category X,and the value greater than separation is counted as non-category X.Then according to the number of category X and non-category X attri?butes,the majority decides the final category prediction.Finally,the experiment is carried out with the tumor tissue attribute data of breast cancer patients in the machine learning library of the University of California,Irvine,USA.It shows that the classifier is sim?ple and easy to implement,with high accuracy.