文献类型：期刊文献

英文题名：Random projection ensemble learning with multiple empirical kernels

作者：Wang, Zhe[1];Jie, Wenbo[1];Chen, Songcan[2];Gao, Daqi[1]

机构：[1]E China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[2]Nanjing Univ Aeronaut & Astronaut, Dept Comp Sci & Engn, Nanjing 210016, Peoples R China

年份：2013

卷号：37

起止页码：388

外文期刊名：KNOWLEDGE-BASED SYSTEMS

收录：;EI(收录号：20124815739885);WOS:【SCI-EXPANDED(收录号:WOS:000313761800035)】；

基金：The authors would like to thank Natural Science Foundations of China under Grant Nos. 60903091, 61170151, 61272198 and 21176077, the Specialized Research Fund for the Doctoral Program of Higher Education under Grant No. 20090074120003, and the Fundamental Research Funds for the Central Universities for partial support.

语种：英文

外文关键词：Multiple kernel learning; Empirical mapping; Random projection; Ensemble classifier; Pattern recognition

摘要：In this paper we propose an effective and efficient random projection ensemble classifier with multiple empirical kernels. For the proposed classifier, we first randomly select a subset from the whole training set and use the subset to construct multiple kernel matrices with different kernels. Then through adopting the eigendecomposition of each kernel matrix, we explicitly map each sample into a feature space and apply the transformed sample into our previous multiple kernel learning framework. Finally, we repeat the above random selection for multiple times and develop a voting ensemble classifier, which is named RPEMEKL. The contributions of the proposed RPEMEKL are: (1) efficiently reducing the computational cost for the eigendecomposition of the kernel matrix due to the smaller size of the kernel matrix; (2) effectively increasing the classification performance due to the diversity generated through different random selections of the subsets: (3) giving an alternative multiple kernel learning from the Empirical Kernel Mapping (EKM) viewpoint, which is different from the traditional Implicit Kernel Mapping (IKM) learning. (C) 2012 Elsevier B.V. All rights reserved.