权核Logistic回归模型的分类和特征选择算法-CEA-计算机工程与应用.PDFVIP

Computer Engineering and Applications 计算机工程与应用 2013 ，49 （9 ） 41 权核Logistic 回归模型的分类和特征选择算法 1 1 2 1 池光辉 ，刘建伟 ，李卫民 ，罗雄麟 1 1 2 1 CHI Guanghui , LIU Jianwei , LI Weimin , LUO Xionglin 1.中国石油大学（北京）自动化研究所，北京 102249 2.上海大学 计算机工程与科学学院，上海 200072 1.Research Institute of Automation, China University of Petroleum, Beijing 102249, China 2.School of Computer Engineering and Science, Shanghai University, Shanghai 200072, China CHI Guanghui, LIU Jianwei, LI Weimin, et al. Classifier and feature selection algorithm by kernel-weighted Logistic regression model. Computer Engineering and Applications, 2013, 49 （9 ）：41-44. Abstract ：Under supervised learning settings, problems that the dimension of the samples is typically larger than the number of samples are often encountered, i.e. many irrelevant features exist. In such case, the approaches that simultaneously achieve sparsely variable selection and better accuracy of classification are more preferable. In this paper, classification and feature slec- tion algorithm based on kernel-weighted nonlinear logistic regression model is proposed. Each diagonal element of the weight diagonal matrix has a value between 0 and 1, which is as a learning parameter determined by optimization procedure, and fast alternative optimization methods are discussed. The proposed methods are tested on ten real-world datasets. The experimental results indicate that the proposed methods show high classification accuracies on these data sets than L 1, L2, Lp norm regulariza- tion classifier algorithm of logistic regression model. Key words ：weighted matrix; logistic regression; feature selection; nonlinear model; kernel function 摘 要：监督学习情况下，经常遇到样例的维数远远大于样本个数的学习情况。此时，样例中存在许多与样例类标签无关 的特征，研究如何同时实现稀疏特征选择并具有更好的分类性能的算法具有优势。提出了基于权核逻辑斯蒂非线性回归 模型的分类和特征选择算法。权对角矩阵