文献类型：期刊文献

英文题名：Group Component Analysis for Multiblock Data: Common and Individual Feature Extraction

作者：Zhou, Guoxu[1,2];Cichocki, Andrzej[3,4,5];Zhang, Yu[6];Mandic, Danilo P.[7]

机构：[1]Guangdong Univ Technol, Sch Automat, Guangzhou 510006, Guangdong, Peoples R China;[2]RIKEN Brain Sci Inst, Lab Adv Brain Signal Proc, Wako, Saitama 3510198, Japan;[3]RIKEN Brain Sci Inst, Wako, Saitama 3510198, Japan;[4]Polish Acad Sci, Syst Res Inst, PL-01447 Warsaw, Poland;[5]Skolkovo Inst Sci & Technol SKOLTECH, Moscow 143025, Russia;[6]East China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[7]Imperial Coll London, Dept Elect & Elect Engn, Commun & Signal Proc Res Grp, London SW7 2BT, England

年份：2016

卷号：27

期号：11

起止页码：2426

外文期刊名：IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS

收录：;EI(收录号：20173404063010);WOS:【SCI-EXPANDED(收录号:WOS:000386940300022)】；

基金：This work was supported in part by the Japan Society for the Promotion of Science through the Grants-in-Aid for Scientific Research Program under Grant 26730125, in part by the National Natural Science Foundation of China under Grant U1201253, Grant 61333013, Grant 61305028, Grant 61273192, and Grant U1401252, in part by the Guangdong Natural Science Foundation under Grant 2014A030308009, and in part by the Guangdong Province Excellent Thesis Foundation under Grant SYBZZXM201316.

语种：英文

外文关键词：Classification; clustering; common and individual feature extraction (CIFE); linked blind source separation (BSS)

摘要：Real-world data are often acquired as a collection of matrices rather than as a single matrix. Such multiblock data are naturally linked and typically share some common features while at the same time exhibiting their own individual features, reflecting the underlying data generation mechanisms. To exploit the linked nature of data, we propose a new framework for common and individual feature extraction (CIFE) which identifies and separates the common and individual features from the multiblock data. Two efficient algorithms termed common orthogonal basis extraction (COBE) are proposed to extract common basis is shared by all data, independent on whether the number of common components is known beforehand. Feature extraction is then performed on the common and individual subspaces separately, by incorporating dimensionality reduction and blind source separation techniques. Comprehensive experimental results on both the synthetic and real-world data demonstrate significant advantages of the proposed CIFE method in comparison with the state-of-the-art.