The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In the fault diagnosis based on support vector machine (SVM), SVM parameters are mostly selected artificially or obtained through experiment time after time, a certain and effective method has not been found. Aiming at this problem, a method optimizing the SVM parameters with Monkey-King genetic algorithm (MKSVM) is presented. In the built model the optimized parameters are used, and the superiority...
A real-time operating system kernel EIRTK with micro-kernel architecture is presented to meet the demands of developing the Internet applications in embedded systems. EIRTK is designed to be a multi-task preemptive kernel based on priority. Task scheduling, synchronization and communication between tasks, timer management, network protocol and other system services are realized in EIRTK to guarantee...
Aiming at the various distribute clustering problems in diffusion model for all data points, providing a new clustering algorithm (CDD) based on the change of density. CDD searches the core point using a typical clustering algorithm (DBSCAN) which based on the density, then calculate the direction, speed and acceleration of density diffused which through analyze the diffusion rule of data sample and...
An adaptive interpolation algorithm is presented based on the Hermite-type interpolation polynomial to improve the limitation of the traditional algorithm for image resizing. The new method for image zooming uses kernel with Hermite-type interpolation spline with local control parameters, which provides a smaller interpolation error and reasonable computing times. Moreover, the method can preserve...
Against the low efficiency of training on large-scale SVM, a reduction approach based on kernel distance clustering is proposed. The kernel distance's formulation is brought in to cluster the highly-dimensioned dataset, and the clustering step will reduce a large amount of unsupport vectors during training, thereby, the training time will decrease. The experiments show that this new training algorithm...
This paper presents high accuracy Nystrom methods (NM) for solving mixed boundary integral equations of scattering problems. We show that errors of the algorithm possess asymptotic expansions with the third power on the mesh parameter. Hence, the accuracy degree of approximations can be greatly improved by extrapolation algorithms (EA). Moreover, a posterior asymptotic error estimate is derived, which...
This paper presents a novel approach to recognize traffic signs using support vector machines and radial Tchebichef moments. More than 3000 real road images were captured by a digital camera under various weather conditions and at different times and locations. After traffic sign is detected from real road images, it is then normalized, and radial Tchebichef moments are computed as the features of...
This paper describes an improved semaphore with policies in Windows operating system. We introduce policies to help operating system kernel select next process (or thread) in the waiting list queue to satisfy. The paper present live policies: first in first out (FIFO), first in last out (FILO), highest priority first out (HPFO), lowest priority first out (LPFO) and random. We discuss the design and...
A new algorithm by using geometric active contour model with the fusion of shape and texture priors to manual segment medical images has been presented in this paper. Then the prior knowledge is merged into active contour model with its contour evolution which is evolved using a genetic algorithm technique. The new method has some advantages over classical level set methods in case of images with...
In this paper, a novel estimation method on Volterra series high-order kernels to nonlinear dynamic system to arbitrary approximation is proposed. On the theoretical basis of kernel function, by the construction of linear space, the issue of solving the Volterra series order kernels is converted to solving the projection of the output of the observation vector in the a sub-space of Hilbert space,...
It is an important method to help doctor's clinical diagnosis that using pattern recognition technology recognizes and counts Urine Sediment's visible component. Harr wavelet feature has good property of distinguish different components, the proposed method using AdaBoost to select a little part typical Harr feature which are taken as input data of SVM. The trained several bi-class SVM classifiers...
A workflow engine with high performance firstly depends on a good structure engine kernel, secondly on the extensibility of function. At present the research of workflow engine focuses on execution of complex business flows, emphasizes application compatibility too much, however, neglects the structure design of engine kernel from an abstract angle. A lightweight workflow engine kernel is designed...
Parallel Finite Difference Time Domain (FDTD) method has been explored over past few years because of the expensive computation needed for its application. And General Purpose Graphics Processing Units (GPGPU), especially Computer Unit Device Architecture (CUDA) model, has been offered an efficient and simple solution. This paper analyzes parallel FDTD method and CUDA architecture, presents a GPU...
This paper applies quadratic Renyi entropy to enterprise financial distress prediction and puts forward a learning algorithm of least squares support vector machines (LS-SVM) based on quadratic Renyi entropy. By respectively analysis and comparison of the algorithm with the traditional LS-SVM, the standard SVM, MLR and BP-ANN, we can see that this algorithm is significantly superior to other algorithms...
Support vector machines (SVMs) have been dominant learning techniques for more than ten years, and mostly applied to supervised learning problems. These years two-class unsupervised and semi-supervised classification algorithms based on bounded C-SVMs, bounded j/-SVMs and Lagrangian SVMs (LSVMs) respectively, which are relaxed to semi-definite programming (SDP), get good classification results. These...
This paper is aimed at the condition monitoring problem of the Power-shift Steering Transmission (PSST), a method of multiple out least squares support vector regression is developed which is applied to prediction of spectrometric oil analysis data. Radial Basis Function (RBF) is used is this algorithm. There are two parameters ?? and ??2. The selection of ?? and ??2 is studied using cross validation...
This paper presents a method to compute liquidity risk of stock market with model of VaR. Firstly, a measure for liquidity is defined, which reflects the volatility of return caused by unite ratio of the position to be liquidated to the tradable shares. Secondly, the density function of the measure for liquidity is estimated with support vector machine, with which the liquidity VaR of stocks is calculated...
To solve the false acceptance problem in speaker recognition, definition of similar centroids among all reference speakers' codebooks and similarity between each two speakers are given. A triple is introduced to mark each centroid with the description of similar centroids. And such triple is used to modify the target function in the fuzzy kernel vector quantization speaker recognition, in which the...
In the process of product intelligent design, product model usually should be established. Generally speaking a complete product model contained a group of attributes, however, in many cases, each attribute needed to be distributed appropriate weight. In this paper the method of extracting attribute sets of product model using product cases was proposed based on attribute reduction in rough set theory,...
With the rapid growth of Internet, the network resource is increasing explosively. Information retrieval is one of main purposes as we browse Internet. At present, there are many retrieval methods and retrieval tools in information retrieval field, users can use all of these avenues to retrieve information. But how to increase the rapidity and precision has become the hotpot in this field. In this...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.