Multi-sample Classification Based On Logistic Regression And Support Vector Machine

As a science of methodology,Statistics plays an important role in both nature science and social science fields.The classification analysis,a method of modern statistical analysis,is an important way for people to understand the world,classifying the objects is helpful for knowing the nature of things.Many statistical methods can be used to solve classification problems,such as k-nearest neighbor,Perceptron,Naive Bayes,Decision tree, Logistic Regression,Support Vector Machine and Neural Network,but for the different classification problems,they have different advantages.This paper will mainly discuss the models,theories,algorithms and applications about the Logistic Regression and Support Vector Machine,and compare their superiority,which can provide the evidences for the choices of multi-classification methods.The structure of this paper:the first chapter is introduction,it introduces the paper’s background and meaning,the paper’s purpose, the selection of models;the second chapter is the theories about models,it contains filling the missing data, factor analysis, logistic regression model,support vector machine model;The fourth chapter is summary and prospect.