在machine learning中，特征降维和特征选择是两个常见的概念，在应用machine learning来解决问题的论文中经常会出现。
　　　 对于这两个概念，很多初学者可能不是很清楚他们的区别。很多人都以为特征降维和特征选择的目的都是使数据的维数降低，所以以为它们是一样的，曾经我也这么以为，这个概念上的误区也就导致了我后面对问题的认识不够深入。后来得到老师的指点才彻底搞清楚了两者的关系，现总结出来与大家分享。(Feature reduction and feature selection are two common concepts in machine learning, which often occur in papers that apply machine learning to solve problems.
For these two concepts, many beginners may not be very clear about their differences. Many people think that the purpose of feature reduction and feature selection is to reduce the dimensionality of data, so they are the same. Once I thought so, the misunderstanding of concept led to the lack of in-depth understanding of the problem behind me. Then we got the guidance of the teacher to make clear the relationship between the two and share it with you.)