Support vector machines (SVMS), a powerful machine method developed from statistical learning and have made significant achievement in some field. Introduced in the early 90's, they led to an explosion of interest...
详细信息
ISBN:
(纸本)0769521789
Support vector machines (SVMS), a powerful machine method developed from statistical learning and have made significant achievement in some field. Introduced in the early 90's, they led to an explosion of interest in machine learning. However, like most machine learning algorithms, they are generally applied using a selected training set classified in advance. With the repaid development of the Internet and telecommunication, huge of information has been produced as digital data format, generally the data is unlabeled. It is impossible to classify the data with one's own hand one by one in many realistic problems, so that the research on unlabeled data classification has been grown. Improvements in databases technology, computing performance and artificial intelligence have contributed to the development of intelligent data analysis. A SVM classifier based on k-means algorithm is presented for the classification of unlabeled data.
Software component frameworks are well known in the commercial business application world and now this technology is being explored with great interest as a way to build large-scale scientific application on parallel ...
详细信息
Software component frameworks are well known in the commercial business application world and now this technology is being explored with great interest as a way to build large-scale scientific application on parallel computers. In the case of grid systems, the current architectural model is based on the emerging Web services framework. We describe progress that has been made on the common component architecture model (CCA) and discuss its success and limitations when applied to problems in grid computing. Our primary conclusion is that a component model fits very well with a services-oriented grid, but the model of composition must allow for a very dynamic (both in space and it time) control of composition. We note that this adds a new dimension to conventional service workflow and it extends the "inversion of control" aspects of must component systems.
We introduce the piecewise-linear Haar (PLHaar) transform, a reversible n-bit to n-bit transform that is based on the Haar wavelet transform. PLHaar is continuous, while all current n-bit to n-bit methods are not, and...
详细信息
We introduce the piecewise-linear Haar (PLHaar) transform, a reversible n-bit to n-bit transform that is based on the Haar wavelet transform. PLHaar is continuous, while all current n-bit to n-bit methods are not, and is therefore uniquely usable with both lossy and lossless methods (e.g. image compression). PLHaar has both integer and continuous (i.e. non-discrete) forms. By keeping the coefficients to n bits PLHaar is particularly suited for use in hardware environments where channel width is limited, such as digital video channels and graphics hardware.
A distributed routing protocol is presented for a wireless ad hoc network that consists of strategic agents. A strategic agent is rational but selfish, and has its own incentive to route traffic for other agents. A me...
详细信息
A distributed routing protocol is presented for a wireless ad hoc network that consists of strategic agents. A strategic agent is rational but selfish, and has its own incentive to route traffic for other agents. A mechanism design (MD) approach is applied and a routing mechanism is designed such that maximizing the benefit of each strategic agent leads to a global optimal system. In this mechanism, an agent accepts payments for forwarding data for other agents if the payments cover their costs incurred by forwarding data. The payment is computed recursively to obtain a cost-efficient and truthful mechanism. The overpayment in normal mechanisms is completely eliminated.
Many communications components are nonlinear and have a peak power or peak amplitude constraint. Nonlinearity generates distortions and thus an appropriate performance measure is the signal-to-noise-and-distortion rat...
详细信息
Many communications components are nonlinear and have a peak power or peak amplitude constraint. Nonlinearity generates distortions and thus an appropriate performance measure is the signal-to-noise-and-distortion ratio (SNDR). In this paper, we are interested in finding the nonlinear mapping that maximizes the SNDR subject to the peak amplitude constraint. The answer is a soft limiter with gain calculated based on the noise power and the probability density function of the input amplitude. We also investigate a bounding relationship between the SNDR and capacity of the nonlinear channel. The results of this paper can be applied for efficient transmission of high peak-to-average power ratio signals such as OFDM or for optimal linearization of nonlinear devices.
Resonance assignment remains one of the hardest stages in RNA tertiary structure elucidation with the use of nuclear magnetic resonance spectroscopy. We propose an evolutionary algorithm being a tool for an automatic ...
详细信息
Resonance assignment remains one of the hardest stages in RNA tertiary structure elucidation with the use of nuclear magnetic resonance spectroscopy. We propose an evolutionary algorithm being a tool for an automatic design of the procedure. NOE pathway, which determines the assignments, is constructed during an analysis of possible connections between resonances within aromatic/anomeric region of 2D-NOESY spectra. Computational tests demonstrate the performance of the genetic algorithm in comparison with the enumerative procedure applied for the experimental and simulated spectral data for RNA molecules.
In this work, we focus on the 3D image reconstruction of buried objects in layered media with sources and receivers located in the top layer. The inversion of scattering data collected by such a partial aperture is in...
详细信息
In this work, we focus on the 3D image reconstruction of buried objects in layered media with sources and receivers located in the top layer. The inversion of scattering data collected by such a partial aperture is in principle more difficult than in the case of a full aperture because the non-uniqueness become more severe. Both the Born iterative method (BIM), and the distorted Born iterative method (DBIM) are used as the inversion methods. In this work, the stabilized biconjugate-gradient fast Fourier transform (BCGS-FFT) method is applied for calculating 3D electromagnetic scattering in general planarly layered media.
This work proposes a fast method for line segment extraction based on chain code differentiation. It is applied to cursive signature recognition of Arabic/Persian. The evaluation method is introduced to obtain a quant...
详细信息
This work proposes a fast method for line segment extraction based on chain code differentiation. It is applied to cursive signature recognition of Arabic/Persian. The evaluation method is introduced to obtain a quantitative value for the recognition rate. The comparative results show the existing differences among the methods in recognition, building time and searching time criteria. The two methods used for comparison are invariant moments and CBLSE method.
A new wideband closed-form Green's function for a HED (horizontal electric dipole) over microstrip geometry is presented. The closed-form is valid over wide range of frequencies, and has the potential application ...
详细信息
A new wideband closed-form Green's function for a HED (horizontal electric dipole) over microstrip geometry is presented. The closed-form is valid over wide range of frequencies, and has the potential application for the time domain analysis of multilayer media by combining it with the MoM. It significantly reduces the computational time, as one only needs to sum few terms instead of repeating the time-consuming Sommerfeld's integrals. It may also be noted that the present method can be easily extended to other kinds of sources and applied to general lossy multilayer media.
This paper deals with neural networks of 8-neighbor for image restoration which are used for gray images. Generally, since comparatively long operation time is needed when a neural network is applied to a certain prob...
详细信息
This paper deals with neural networks of 8-neighbor for image restoration which are used for gray images. Generally, since comparatively long operation time is needed when a neural network is applied to a certain problem, it is important to estimate the convergence rate of the neural network in order to evaluate the operation time. In this paper, the theoretical convergence rate of the neural network for image restoration is derived both for the case where an image consists of not many pixels and for the case where an image consists of many pixels. Furthermore, it is shown that the two convergence rates are appropriate through a numerical experiment.
暂无评论