Yang T.-NSHENG-DE WANG2009-03-042018-07-062009-03-042018-07-06199901678655http://scholars.lib.ntu.edu.tw/handle/123456789/352564http://ntur.lib.ntu.edu.tw/bitstream/246246/142280/1/17.pdfhttps://www.scopus.com/inward/record.uri?eid=2-s2.0-0032594808&doi=10.1016%2fS0167-8655%2899%2900060-4&partnerID=40&md5=6b8ea3e069045391a7073fa8a5a2ce6dIn this paper, we address the issues related to the design of fuzzy robust principal component analysis (FRPCA) algorithms. The design of robust principal component analysis has been studied in the literature of statistics for over two decades. More recently Xu and Yuille proposed a family of online robust principal component analysis based on statistical physics approach. We extend Xu and Yuille's objective function by using fuzzy membership and derive improved algorithms that can extract the appropriate principal components from the spoiled data set. The difficulty of selecting an appropriate hard threshold in Xu and Yuille's approach is alleviated by replacing the threshold by an automatically selected soft threshold in FRPCA. Artificially generated data sets are used to evaluate the performance of various PCA algorithms. © 1999 Elsevier Science B.V.application/pdf193730 bytesapplication/pdfen-USFuzzy theory; Neural networks; Noise clustering; Principal component analysis; Robust algorithmAlgorithms; Fuzzy sets; Membership functions; Neural networks; Statistical methods; Noise clustering; Principal component analysis; Pattern recognitionjournal article10.1016/S0167-8655(99)00060-42-s2.0-0032594808http://ntur.lib.ntu.edu.tw/bitstream/246246/142280/1/17.pdf