The Influence of Polynomial Order in Logistic Regression on Decision Boundary

2019 3rd International Workshop on Renewable Energy and Development

能源学;生态环境科学

Wan, Xing^1

Leshan Vocational and Technical College, Leshan, Sichuan

614000, China^1

关键词: Approximate optimal solutions;    Binary classification problems;    Decision boundary;    Logistic regression algorithms;    Logistic regressions;    Machine learning problem;    Nonlinear problems;    Polynomial regression;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/267/4/042077/pdf
DOI  :  10.1088/1755-1315/267/4/042077

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

In machine learning problems, polynomial logistic regression algorithms are often used to classify data. Compared to linear regression, polynomial regression can not only deal with linear problems, but also deal with nonlinear problems. In the polynomial logistic regression algorithm, the polynomial order has a certain influence on the classification effect. This paper studies the influence of the polynomial order on the binary decision boundary in binary classification problem. By choosing different parameter values, an approximate optimal solution can be found.