基于数据降维的机器学习分类应用问题探讨

Discussion on Application of Machine Learning Classification Based on Data Dimension Reduction

本文以当前机器学习分类应用问题作为切入点,简述样本归类不清、计算量较大等不足,再以此为基础,重点介绍两类数据降维方式,并以模拟实验证明随机森林法、K近邻算法的优势,以期通过分析明晰理论,为后续机器学习工作提供参考。

based on the current application of machine learning classification problem as the breakthrough point,the paper sample classification is not clear,such as large amount of calculation,and on this basis,the emphasis on the two types of data dimension reduction method,and the simulation experiments prove the advantage of random forest law,K neighbor algorithm,in order to through the analysis of a clear theory,provide a reference for the follow-up work of machine learning.