检索结果-内蒙古大学图书馆

2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing, MLSP 2006

作者： Shen, Hao Hüper, Knut Systems Engineering and Complex Systems Research Program National ICT Australia Canberra ACT 2601 Australia Department of Information Engineering RSISE Australian National University Canberra ACT 0200 Australia

ISBN: (纸本)1424406560

Independent Component Analysis (ICA) can be studied from different angles. The performance of ICA algorithms significantly depends on the choice of the contrast function and the optimisation algorithm used in obtaining the demixing matrix. In this paper we focus on the standard linear ICA problem from an algorithmic point of view. It is well known that after a pre-whitening process, linear ICA problem can be solved via an optimisation approach on a suitable manifold. FastICA is one prominent linear ICA algorithm for solving the so-called one-unit ICA problem, which was recently shown by the authors to be an approximate Newton's method on the real projective space. To extract multiple components in parallel, in this paper, we propose an approximate Newton-like ICA algorithm on the orthogonal group. The local quadratic convergence properties are discussed. The performance of the proposed algorithms is compared with several existing parallel ICA algorithms by numerical experiments. © 2006 IEEE.

关键词： Independent component analysis

来源：评论

学校读者我要写书评

暂无评论

An efficiency evaluation model of EDI in a value added network: The influence of user confidence on risk

An efficiency evaluation model of EDI in a value added netwo...

引用

11th International Conference of the Association information and Management, AIM 2006

作者： Baile, Serge Research Center on Organizational Development and Information Systems Business Administration Institute IAE-University of Toulouse 1 France Program in Information Systems Management and Engineering Toulouse Business School-Groupe ESC Toulouse Financial Research Group France

ISBN: (纸本)9783885791867

The evaluation of the performance of information technology and information systems has, over the past three decades, been the object of numerous studies using a wide range of models. Few of them have focused on the study of efficiency, and even fewer still on the development of a specific research model. This paper helps to fill this gap by making a contribution towards formalizing a specific, original theoretical framework and conceptual model, one based on user confidence in ad hoc EDI technology and the risks associated with its use. The model is tested using data obtained from users of a value added network in the car insurance sector. The results of the data analysis, obtained through a series of principal component analysis methods with a Promax rotation and an Amos structural equation analysis, confirm the existence, on the one hand, of an optimal structure of EDI efficiency evaluation factors, and on the other hand, of the relevance of a data model that establishes a hierarchy aimed at explaining risk.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Novelty-based incremental document clustering for on-line documents 22

Novelty-based incremental document clustering for on-line do...

引用

22nd International Conference on Data engineering Workshops, ICDEW 2006

作者： Khy, Sophoin Ishikawa, Yoshiharu Kitagawa, Hiroyuki Master's Program in Science and Engineering Tsukuba Ibaraki Japan Graduate School of Systems and Information Engineering Tsukuba Ibaraki Japan Center for Computational Sciences University of Tsukuba Tsukuba Ibaraki Japan

ISBN: (纸本)0769525717

Document clustering has been used as a core technique in managing vast amount of data and providing needed information. In on-line environments, generally new information gains more interests than old one. Traditional clustering focuses on grouping similar documents into clusters by treating each document with equal weight. We proposed a novelty-based incremental clustering method for on-line documents that has biases on recent documents. In the clustering method, the notion of 'novelty' is incorporated into a similarity function and a clustering method, a variant of the K-means method, is proposed. We examine the efficiency and behaviors of the method by experiments. © 2006 IEEE.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Workshop on imaging science development for cancer prevention and preemption

引用

Cancer Biomarkers 2007年第1期3卷 1-33页

作者： Kelloff, Gary J. Sullivan, Daniel C. Baker, Houston Clarke, Lawrence P. Nordstrom, Robert Tatum, James L. Dorfman, Gary S. Jacobs, Paula Berg, Christine D. Pomper, Martin G. Birrer, Michael J. Tempero, Margaret Higley, Howard R. Petty, Brenda Gumbs Sigman, Caroline C. Maley, Carlo Sharma, Prateek Wax, Adam Ginsberg, Gregory G. Dannenberg, Andrew J. Hawk, Ernest T. Messing, Edward M. Grossman, H. Barton Harisinghani, Mukesh Bigio, Irving J. Griebel, Donna Henson, Donald E. Fabian, Carol J. Ferrara, Katherine Fantini, Sergio Schnall, Mitchell D. Zujewski, Jo Anne Hayes, Wendy Klein, Eric A. DeMarzo, Angelo Ocak, Iclal Ketterling, Jeffrey A. Tempany, Clare Shtern, Faina Parnes, Howard L. Gomez, Jorge Srivastava, Sudhir Szabo, Eva Lam, Stephen Seibel, Eric J. Massion, Pierre McLennan, Geoffrey Cleary, Kevin Suh, Robert Burt, Randall W. Pfeiffer, Ruth M. Hoffman, John M. Roy, Hemant K. Wang, Thomas Limburg, Paul J. El-Deiry, Wafik S. Papadimitrakopoulou, Vali Hittelman, Walter N. MacAulay, Calum Veltri, Robert W. Solomonzz, Diane Jeronimo, Jose Richards-Kortum, Rebecca Johnson, Karen A. Viner, Jaye L. Stratton, Steven P. Rajadhyaksha, Milind Dhawan, Atam NIH/NCI/DCTD Cancer Imaging Program Bethesda MD United States NIH/NCI/DCTD Cancer Imaging Program NCI-Frederick Frederick MD United States NIH/NCI/DCP Early Detection Research Group Bethesda MD United States The Johns Hopkins University School of Medicine Departments of Radiology Pharmacology and Oncology Division of Neuroradiology Baltimore MD United States NIH/NCI/CCR Cell and Cancer Biology Branch Bethesda MD United States University of California San Francisco Department of Medicine Division of Medical Oncology San Francisco CA United States CCS Associates Inc. Mountain View CA United States Wistar Institute Molecular and Cellular Oncogenesis Program Philadelphia PA United States University of Kansas School of Medicine VA Hospital Kansas City MO United States Duke University Department of Biomedical Engineering Medical Physics Program Durham NC United States University of Pennsylvania School of Medicine Department of Gastroenterology Philadelphia PA United States New York Presbyterian Hospital Weill Medical College of Cornell University New York NY United States NIH/NCI/OD Office of Centers for Training and Resources Bethesda MD United States University of Rochester Medical Center Department of Urology Rochester NY United States University of Texas MD Anderson Cancer Center Department of Urology Houston TX United States Massachusetts General Hospital Abdominal Imaging/Interventional Radiology Boston MA United States Boston University Biomedical Engineering Electrical and Computer Engineering and Physics Boston MA United States NIH/NCI/DCP Gastrointestinal and Other Cancers Research Group Bethesda MD United States The George Washington University Cancer Institute Office of Cancer Prevention and Control Washington DC United States University of Kansas Cancer Center Clinical Oncology Kansas City KS United States University of California Davis Department of Biomedical Engineering Davis CA United States Tufts University De

The article provides information on a workshop on the development of imaging systems for cancer prevention and preemption convened by the U.S. National Cancer Institute (NCI) Cancer Imaging program held in Gaithersburg, Maryland. Doctor Katherine Ferrara of the University of California discussed the function of ultrasound in breast tumor imaging. Ways to close the gap between development and application of imaging approaches were tackled in a roundtable discussion on cervix and skin cancer.

关键词： FORUMS (Discussion & debate) IMAGING systems in medicine CANCER prevention

来源：评论

学校读者我要写书评

暂无评论

Newton-Like Methods for Parallel Independent Component Analysis

Newton-Like Methods for Parallel Independent Component Analy...

引用

IEEE Workshop on Machine Learning for Signal Processing

作者： Hao Shen Knut Huper Systems Engineering and Complex Systems Research Program National ICT Australia Canberra Australia Department of Information Engineering The Australian National University Canberra

Independent component analysis (ICA) can be studied from different angles. The performance of ICA algorithms significantly depends on the choice of the contrast function and the optimisation algorithm used in obtaining the demixing matrix. In this paper we focus on the standard linear ICA problem from an algorithmic point of view. It is well known that after a pre-whitening process, linear ICA problem can be solved via an optimisation approach on a suitable manifold. FastICA is one prominent linear ICA algorithm for solving the so-called one-unit ICA problem, which was recently shown by the authors to be an approximate Newton's method on the real projective space. To extract multiple components in parallel, in this paper, we propose an approximate ***-like ICA algorithm on the orthogonal group. The local quadratic convergence properties are discussed. The performance of the proposed algorithms is compared with several existing parallel ICA algorithms by numerical experiments..

关键词： Independent component analysis Signal processing algorithms Convergence Newton method Source separation Biomedical signal processing Australia Council systems engineering and theory Blind source separation Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Local Convergence Properties of Fastica and Some Generalisations

Local Convergence Properties of Fastica and Some Generalisat...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： K. Huper Hao Shen A.-K. Seghouane Systems Engineering and Complex Systems Research Program National ICT Australia Cambridge Research Laboratory Canberra ACT Australia Department of Information Engineering Research School of Information Sciences and Engineering Australian National University Canberra ACT Australia

In recent years, algorithms to perform independent component analysis in blind identification, localisation of sources or more general in data analysis have been developed. Prominent example certainly is the socalled FastICA algorithms from the Finnish school. In this paper we will generalise the FastICA algorithm considered as a discrete dynamical system on the unit sphere to the case where all units converge simultaneously, i.e., we consider some kind of parallel FastICA algorithm living on the orthogonal group. In addition we present a local convergence analysis for the algorithms proposed in this paper building on earlier work. It turns out that one can treat these type of algorithms in a similar manner as the Rayleigh quotient iteration, well known in numerical linear algebra, i.e. considering the algorithm as a discrete dynamical system on a suitable manifold. The algorithms presented here are compared by several numerical experiments and simulations

关键词： Convergence Signal processing algorithms Independent component analysis Data engineering Educational institutions Source separation Australia Council systems engineering and theory Laboratories Data analysis

来源：评论

学校读者我要写书评

暂无评论

Smooth interpolation of orientation by rolling and wrapping for robot motion planning

Smooth interpolation of orientation by rolling and wrapping ...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Yueshi Shen K. Huper F.S. Leite Department of Information Engineering Australian National University Canberra Australia Systems Engineering and Complex Systems Program NICTA Canberra Australia Department of Mathematics and Institute of Systems and Robotics University of Coimbra Coimbra Portugal

This paper investigates a novel procedure to calculate smooth interpolation curves of the rotation group SO 3 , which is commonly considered as the standard representation of rigid-body's orientations. The algorithm is a combination of rolling and wrapping with the pull back/push forward technique. One remarkable advantage of this approach is that interpolation curves are given in closed form, which brings convenience for implementations on real-time control systems. A numerical example along with some visualization results is presented as well

关键词： Interpolation Wrapping Robot motion Motion planning Robot kinematics Orbital robotics Australia Council Forward contracts systems engineering and theory Mathematics

来源：评论

学校读者我要写书评

暂无评论

Geometric Splines and Interpolation on S2: Numerical Experiments

Geometric Splines and Interpolation on S2: Numerical Experim...

引用

IEEE Conference on Decision and Control

作者： K. Huper Y. Shen F. Silva Leite National ICT Australia Canberra Research Laboratory Systems Engineering and Complex Systems Program Canberra Australia Department of Information Engineering Research School of Information Sciences and Engineering Australian National University Canberra Australia Department of Mathematics and the Institute of Systems and Robotics University of Coimbra Coimbra Portugal

ISBN: (纸本)1424401704;9781424401703

Several different procedures are presented to produce smooth interpolating curves on the two-sphere S 2 . The first class of methods is a combination of the pull back/push forward technique with unrolling data from S 2 into a tangent plane, solving there the interpolation problem, and then wrapping the resulting interpolation curve back to the manifold. The second method results from converting a variational problem into a finite dimensional optimisation problem by a proper discretisation process. It turns out that the resulting curves look very similar. The main difference though is that the first approach gives closed form solutions to the interpolation problem, whereas the second method results in a finite number of points. These points then require further treatment, e.g., one could connect them by geodesic arcs, i.e., by great circle segments, to get an approximate solution to the variational problem. Although the result would not be smooth, it seems to be the best that one can get if the discretisation process is combined with a sufficiently cheap interpolation procedure

关键词： Interpolation Manifolds USA Councils Wrapping Optimization methods Closed-form solution Optimal control Robot kinematics Orbital robotics Path planning

来源：评论

学校读者我要写书评

暂无评论

Predicting Probability of Mortality in the Neonatal Intensive Care Unit

Predicting Probability of Mortality in the Neonatal Intensiv...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Dajie Zhou Monique Frize Program of Systems Science University of Ottawa Ottawa ONT Canada Department of Systems and Computer Engineering Carleton University and the School of Information Technology and Engineering University of Ottawa Ottawa ONT Canada

Artificial neural networks can be trained to predict outcomes in a neonatal intensive care unit (NICU). This paper expands on past research and shows that neural networks trained by the maximum likelihood estimation criterion will approximate the `a posteriori probability' of NICU mortality. A gradient ascent method for the weight update of three-layer feed-forward neural networks was derived. The neural networks were trained on NICU data and the results were evaluated by performance measurement techniques, such as the Receiver Operating Characteristic Curve and the Hosmer-Lemeshow test. The resulting models applied as mortality prognostic screening tools are presented

关键词： Pediatrics Neural networks Artificial neural networks Maximum likelihood estimation Measurement Testing Feedforward systems Feedforward neural networks Resource management Cities and towns

来源：评论

学校读者我要写书评

暂无评论

Nonlinear System Identification Based on B-Spline Neural Network and Modified Particle Swarm Optimization

Nonlinear System Identification Based on B-Spline Neural Net...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： L. dos Santos Coelho R.A. Krohling Production and Systems Engineering Graduate Program Automation and Systems Laboratory PPGEPS/CCET/PUCPR Pontifical Catholic University of Paraná Parana Brazil Chair for Control Systems Engineering RST Faculty of Electrical Engineering and Information Technology University of Dortmund Dortmund Germany

Artificial neural networks, in particular, feedforward multilayer networks and basis function networks, have gradually established themselves as a usual tool in approximating complex nonlinear systems. B-spline networks, a type of basis function neural network, are normally trained by gradient-based methods, which may fall into local minima during the learning phase. In order to overcome the drawbacks encountered by conventional learning methods, particle swarm optimization - a swarm intelligence methodology - can provide a stochastic global search of B-spline networks for nonlinear system identification. In this paper, a modified particle swarm optimization algorithm using Gaussian and Cauchy probability distributions are applied to adjust the control points of B-spline neural networks. Simulation results for the identification of Rossler systems are provided and demonstrate the effectiveness and robustness of the proposed identification scheme.

关键词： Nonlinear systems Spline Neural networks Particle swarm optimization Artificial neural networks Multi-layer neural network Learning systems Stochastic systems Probability distribution Robustness

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：