摘要：
Large structure modal parameter estimation has always been an important research content. Its core content is to obtain the eigenvalues of large-scale structural system. Based on the environmental excitaion, to obtain high precision system free response is particularly important. This paper presents a random reduction de-noising-ARMA method(RDT-ARMA method). By analyzing the characteristics of free response root and imaginary part under the condition of health and damage, the noise modal of free response obtained by stochastic reduction method can be effectively eliminated to obtain a more realistic system free response signal. Based on this, the modal of large structural system is identified by ARMA method. And through the actual acquisition of the WSN signal processing measured acceleration, indicating the effectiveness of the proposed method can be proved.
关键词：
Multiview data;Partial least squares;Regression
摘要：
In practice, multiple distinct features are need to comprehensively analyze complex samples. In machine learning, data set obtained with a feature extractor is referred as a view. Most of data used in practics are collected with various feature extractors. It is practical to assume that an individual view is unlikely to be sufficient for effective analyzing the property of the sample. Therefore, integration of multiview information is both valuable and necessary. But, traditional partial least squares is proposed for single view high dimensional data modeling,which is invalid for multiview data. In this paper, multiveiw partial least squares is proposed. This model finds a series of direction vectors which guarantee covariance between response and weighted component reach maximum as well as pairwise correlation of component. We then proposed an algorithm for multiview partial least squares. Convergence and bound discussion are also given. Experiments demonstrate that proposed multiview partial least squares is an effective and promising algorithm for practical applications.
摘要：
The partial least squares (PLS) method is designed for prediction problems when the number of predictors is larger than the number of training samples. PLS is based on latent components which are linear combinations of the original predictors, it automatically employs all predictors regardless of their relevance. This strategy will potentially degrade its performance, and make the obtained coefficients lack interpretability. Then, several sparse PLS (SPLS) methods are proposed to simultaneously conduct prediction and variable selection via sparsely combining original predictors. However, if information bleed across different components, common variables shared by these components should be selected with successive loadings. To address this issue, we propose a new SPLS model mixed-norm PLS (MNPLS) to select common variables during each deflation in this paper. More specifically, we introduced the l(2,1) norm to the direction matrix and then developed the corresponding solution to MNPLS. Moreover, we also conducted convergence analysis to mathematically support the proposed MNPLS. Experiments are conducted on four real datasets, experimental results verified our theoretical analysis, and also demonstrated that our MNPLS method can generally outperform the standard PLS and other existing methods in variable selection and prediction. (C) 2016 Elsevier B.V. All rights reserved.
摘要：
Calibration model transfer is an important issue in infrared spectra analysis. For identical sample, spectra collected with master and slave spectrometers share same components. In the sense of mathematics, they share same basis. If the basis and corresponding coefficient matrices can be obtained, the model transfer can be efficiently realized. On the other hand, the performance of calibration model transfer method will degrade if there are outliers and noise in samples. In this paper, a robust calibration transfer model is proposed. Cauchy estimator are employed to learn same basis shared by master and slave spectra robustly. Transformation matrix can be calculated with the two corresponding coefficient matrices. Slave testing spectra are represented with the common basis and corresponding coefficients are then transferred using the transformation matrix. The slave testing spectra can be transferred using common basis and the corrected coefficients. The convergence property and bound of proposed model are also discussed. Extensive experiments are conducted, experimental results demonstrate that our robust calibration transfer model can generally outperform the existing methods.
关键词：
Infrared spectra;Light scattering;Regularized least square;RMSC
摘要：
As an efficient method for spectra correction, multivariate scatter correction (MSC) has recently received considerable attention due to the precision improvement of processed data. In general, the spectra approximate mean spectrum S in least square framework. Unfortunately, the existing MSC methods have a limited capability in nonlinear component modeling. In this paper, we propose regularized multivariate scatter correction (RMSC), which has taken nonlinear components into MSC model as well as regularization function for the weight vector w. The weighted sum of mappings of observed spectrum is used to approximate the mean spectrum. By using gradient projection sparse representation, vector w is obtained for RMSC. Results show a substantial decrease in Root Mean Square Error of Prediction of quantitative analysis and improvement in classification precision. Crown Copyright (c) 2013 Published by Elsevier B.V. All rights reserved.
会议论文集名称：
Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)
摘要：
Variable selection has been extensively studied in linear regression and classification models. Most of these models assume that the input variables are noise free, the response variables are corrupted by Gaussian noise. In this paper, we discuss the variable selection problem assuming that both input variables and response variables are corrupted by Gaussian noise. We analyze the prediction error when augment one related noise variable. We show that the prediction error always decrease when more variable were employed for prediction when the joint distribution of variables are known. Based on this analysis, in sense of mean square error, the optimal variable selection can be obtained. We found that the results is very different from the matching pursuit algorithm(MP), which is widely used in variable selection problems.
期刊：
Proceedings of the IASTED International Conference on Signal Processing, Pattern Recognition and Applications, SPPRA 2013,2013年:215-221
通讯作者：
Mou, Y.(my515@yahoo.cn)
作者机构：
[Jiang, Xiubao; Mou, Yi; Yu, Shujian; Ou, Weihua; You, Xinge] Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan 430074, China;[Zhou, Long] School of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430023, China
摘要：
A new baseline correction algorithm for spectral signal based on sparse representation is proposed. Firstly, utilizing the training sample to obtain the dictionaries of both baseline and spectrum;Secondly, establishing sparse representation model of spectral signal;thirdly, employing OM-P algorithm to calculate the representation coefficients of spectral signal and finally, obtaining the spectral baseline from representation coefficients which are corresponded to the baseline dictionary. Then, the spectra baseline correction is completed by removing the baseline from original observed spectrum. Contrast experiment and quantitative analysis of corrected spectral signals are conducted and results show the highly efficiency and accuracy of the proposed algorithm.
作者机构：
[周龙; 尤新革] Department of Electrical and Information Engineering, Wuhan Polytechnic University, Wuhan 430023, China;[牟怿] School of Engineering, Anhui Science and Technology University, Fengyang 233100, Anhui, China
通讯机构：
Department of Electrical and Information Engineering, Wuhan Polytechnic University, China
作者机构：
[刘纯利; 牟怿] Engineering School, Anhui Science and Technology University, Fengyang 233100, Anhui, China;[周龙] Department of Electrical and Information Engineering, Wuhan Polytechnic University, Wuhan 430023, China
通讯机构：
Engineering School, Anhui Science and Technology University, China
作者机构：
[周龙; 牟怿] Department of Electrical and Information Engineering, Wuhan Polytechnic University, Wuhan 430023, China;[黄凌霄] School of Mathematics and Computer Science, Ningxia University, Yinchuan 750021, China
通讯机构：
Department of Electrical and Information Engineering, Wuhan Polytechnic University, China
