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.
BP Neural Networks is one of the most popular tools in the analysis of stock data. Recent research activities in Pattern Matching indicate that Pattern Matching just simplify the complexity of stock trend prediction and provide a simple but effective way for the stock market prediction. This paper analysis the theory of BP Neural Networks and Pattern Matching, proposes a method for combining these...
We here propose implementation of some architectures of artificial neural networks as a potential solution of that problem prediction in electronics based on short time series. Examples are given related to predictions for verification of Moore's law in modern electronic developments.
According to analyzing the datum of the leakage currents from both perfect and faulty insulators that are at the state of various dirtiness, humidity, and temperature, and are applied non-sinusoidal voltages at different virtual values, three results are obtained. First, the bispectrum of the leakage current from contaminated insulator is mostly related to insulator; not to dirtiness. Bispectrum is...
The phase space reconstruction and artificial neural networks (ANN) coupled model is developed for flow forecasting in consideration of chaotic property and nonlinearity of flow series. 50 years of monthly flow data from 1950 to 1999 in Yichang hydrologic station is used for parameter calibration, and 4 years of the data from 2000 to 2003 is used for model validation. The result shows it has high...
In this paper, the algorithm combined the construction methods of immune strategies and immune operator, better solved the degradation phenomenon appeared in the algorithm, and the convergence speed has been improved significantly. Comparing with the differential evolution algorithm, adding differential evolution operator in the immune algorithm can increase antigen recognition, memory function and...
Distributed systems have been developing rapidly in the past few years and their automatic control is a real challenge being a very active research field. In order to assure the load balancing and to optimize the resource utilization, a distributed system is using different software components, such as management tools, schedulers or monitoring tools. Considering the prediction of future behavior...
Wavelet neural network (WNN) trained by unscented Kalman filter (UKF) has many merits of fast convergent rate and small prediction error without computing the Jacobian matrix. Based on this, an improved UKF is introduced into the parameters estimation for WNN. The algorithm uses an unscented transform (UT) based on minimal skew simplex Sigma point sampling strategy in the frame of Kalman filter, which...
In this work, we present a neural network (NN) based method designed for 3D rigid-body registration of FMRI time series, which relies on a limited number of Fourier coefficients of the images to be aligned. These coefficients, which are comprised in a small cubic neighborhood located at the first octant of a 3D Fourier space (including the DC component), are then fed into six NN during the learning...
In the intensive care unit (ICU), prompt therapeutic intervention to hypotensive episodes (HEs) is a critical task. Advance alerts that can prospectively identify patients at risk of developing an HE in the next few hours would be of considerable clinical value. In this study, we developed an automated, artificial neural network HE predictor based on heart rate and blood pressure time series from...
In this paper, we present the use of higher order neural networks for the prediction of speech signal. Various neural network structure have been used for our experiments these include the functional link neural network and pi-sigma neural network. Extensive experimentation is carried out to evaluate the performance of the higher order networks on the speech prediction platform. Simulation results...
A novel application to the optimization of artificial neural networks (ANNs) is presented in this paper. Here, the weight and architecture optimization of ANNs can be formulated as a mixed-integer optimization problem. And then a mixed-integer evolutionary algorithm (Mixed-Integer Hybrid Differential Evolution, MIHDE) is used to optimize the ANN. Finally, the optimized ANN is applied to the prediction...
Spiking neural networks (SNNs) are considered as the third generation of neural network models. In this paper, we proposed a supervised learning rule for SNNs that can cope with neurons, which can fire multiple spikes. The proposed supervised learning rule is based on the linear algebra method. We implemented the supervised learning rule for approximating the output to the goal time series. Simulation...
As is known to all, the neutral network has made a great progress in many fields. But due to some strict theoretical system, there are still many defaults in practical application. In this paper, we present an active learning artificial neural network (ALANN). The key issue of this kind of approach is what information can be analysis and forecast about time series(TS). However, the parameters of ALANN...
The prediction of financial time series is a very complicated process. An initial look at financial time series gives the impression that they are random in nature. If true, this would make the forecast, and therefore the trading, of such series exceptionally difficult. The efficient market hypothesis states that the current price contains all available information in the market. This leads to the...
The BP neural network has proven robust even for complex nonlinear problems. However, its high performance results are attained at the expense of a long training time to adjust the network parameters, which can be discouraging in many real-world applications. Even on relatively simple problems, standard BP often requires a lengthy training process in which the complete set of training examples is...
The forecasting to a nation's freight and its turnover is an important thing for the development of transportation industry and even national economy. Based on the analysis using Backpropagation Algorithm in artificial neural network and time series forecasting methods in Microsoft Excel to China's transportation volume from 1995 to 2007, the paper try to find the algorithm to effectively predict...
Abstract-This paper presents a new approach to recognize and predict succedent epileptic seizures by using single-channel electroencephalogram (EEG) analysis. Eight channels of EEG from each patient of the seven consenting patients with generalized epilepsy were collected in Epilepsy Center of Xijing Hospital. The raw EEGs were decomposed by the algorithm of empirical mode decomposition (EMD), the...
In the analysis of predicting power load forecasting based on least squares neural network, the instability of the time series could lead to decrease of prediction accuracy. On the other hand,neural network and chaos theories parameters must be carefully predetermined in establishing an efficient model. In order to solve the problems mentioned above, in this paper, the neural network and chaos theory...
To address the problem of approximation and prediction of complex time-varying system, this paper proposes a parallel process neural networks predication method based on general process neural networks models. Firstly, the whole time-varying process is divided into several small time intervals; then, the process neural networks are constructed respectively in the small time intervals to disperse the...
The traditional stationary network traffic model (ARIMA) is incapable of describing non-stationary characteristics. In the process of predicting, the accuracy will weaken with the increase of step. As a non-stationary network traffic model, NN (neural network) could make up for the defect of stationary model, which can not describe the non-stationary qualities of the network traffic. However, the...
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.