文档介绍：第 36 卷第 4 期自动化学报 Vol. 36, No. 4
2010 年 4 月 ACTA AUTOMATICA SINICA April, 2010
基于增量核主成分分析的数据流在线分类框架
吴枫 1 仲妍 1 吴泉源 1
摘要核主成分分析(Kernel ponent analysis, KPCA) 是一种非线性降维工具, 在降低数据流分类处理量
方面发挥着积极作用. 然而, 由于复杂性太高, 导致 KPCA 的降维能力有限. 为此, 本文给出了一种增量核主成分分析算法
(Incremental KPCA for dimensionality-reduction, IKDR), 该算法在每步迭代估计中只需线性内存开销, 大大降低了复杂性.
在 IKDR 的基础上, 结合 BP (Back propagation) 神经网络提出了数据流在线分类框架: IKOCFrame (Online classification
frame based on IKDR). 通过一系列真实和人工数据集上的实验, 检验了 IKDR 算法的收敛性, 并且验证了 IKOCFrame 相
对于同类基于成分分析的分类算法的优越性.
关键词降维技术, 数据流分类, 增量核主成分分析, 独立成分分析
DOI .
Online Classification Framework for Data Stream Based on Incremental
Kernel ponent Analysis
WU Feng1 ZHONG Yan1 WU Quan-Yuan1
Abstract Kernel ponent analysis (KPCA) has been suggested for various data stream classification tasks
requiring a nonlinear transformation scheme to reduce dimensions. However, the dimensionality reduction ability is
restricted because of its plexity. Therefore this paper proposes an incremental kernel ponent
analysis algorithm: IKDR, which iteratively estimates the kernel ponents with only linear order storage
complexity per iteration. On the basis of IKDR, this paper proposes an online classification framework for data stream:
IKOCFrame. Extensive experiments on real and artificial datasets validate the convergence of IKDR and confirm the
superiority of IKOCFrame