Focusing on the data with high dimensions and nonlinearity
in pattern recognition in high dimensional observation space
an improved manifold learning algorithm is introduced
and a new approach is proposed by combining adaptive local linear embedding(ALLE)and recursively applying normalized cut algorithm(RANCA).The adaptive local linear embedding algorithm is employed for nonlinear dimension reduction of original dataset
then recursively applying normalized cut algorithm is used in clustering of low dimensional data.The simulation results of three UCI standard datasets show that the new method can map high-dimensional data into low-dimensional intrinsic space successfully
solves the more dependence on the structure of datasets in the traditional methods
and the classification accuracy and robustness of spectral clustering algorithm are remarkably improved. The experiment results on tennessee-eastman process(TEP)also demonstrate the feasibility and effectiveness of the new method in fault pattern recognition.
关键词
Keywords
references
FRANK P M. Analytical and qualitative model-based fault diagnosis-a survey and some new results [J]. European Journal of Control, 1996, 2(1): 6-28.
WANG Na, DU Haifeng, ZHUANG Jian, et al. Spectral clustering method based on network segmentation used in fault diagnosis [J]. Chinese Mechanical Engineering, 2008, 44(10):228-233.
SEUNG H S, DANIEL D L. The manifold ways of perception [J]. Science, 2000, 290(12): 2268-2269.
ROWEIS S T, SAUL L K. Nonlinear dimensionality reduction by local linear embedding [J]. Science, 2000, 290: 2323-2326.
SAUL L K, ROWEIS S T. Think globally, fit locally: unsupervised learning of nonlinear manifolds [J]. Journal of Machine Learning Research, 2003, 4(12): 119-155.
SHI J, MALIK J. Normalized cuts and image segmentation [J]. IEEE Transactions on PAMI, 2000, 22(8): 888-905.
GOLUB G H, VANLOAN C F. Matrix computations [M]. Beijing,China:China Machine Press, 2001.
NEWMAN D J, HETTICH S, BLAKE C L. UCI repository of machine learning databases[S/OL]. Irvine, CA: University of California, Department of Information and Computer Science. [2009-04-16]. http:∥www.ics.uci.edu~mlearn/MLRepository.html.