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.
A simple and robust approach for islanding detection is introduced in this paper. The proposed approach detects islanding using the transient signals. The three phases' currents seen at the DG terminals are combined into one modal signal that fully represents the system. The feature vector is extracted from the selected modal current signal utilizing discrete wavelet transform. The extracted feature...
This article proposes an audio-visual objective quality model, which is adaptive to the content changes. The major disadvantage of most existing models is that there is no unified model to evaluate the quality according to the content. So we firstly extracted the audio and video features and the correlation of the audio and video. Then we used BP neutral network to build the model, and used the subjective...
Two years (2007–2008) of observational particulate matter (PM) data from ten ground-stations over eastern China, as well as collocated Aerosol Optical Depth (hereafter called AOD) derived from Moderate Resolution Imaging Spectroradiometer (MODIS), in conjunction with meteorological parameters, such as planetary boundary layer height (PBLH), relative humid (RH), temperature (TEMP), wind speed (WS),...
Obtaining the quantitative positioning space-based demographic and socio-economic information has the significance on assessing resources, environment and disaster. This paper presents a dynamic modeling method for rural GDP statistics data spatialization based on neural network Selecting Fangshan District in Beijing, China as the study area and taking villages as studying unit, this paper analyzes...
The data mining technology is more and more widely used in the telecom industry. But telecom data set always includes instances with missing values. Besides, many data mining models are sensitive for the missing value and distortion. Estimating missing values becomes an inherent problem. To address the problem, A prediction method is proposed for the missing value based on the BP neural network and...
Input variables selection plays a critical role in data-driven modelling, especially for complex systems with high dimensionality between the input/output space. In this paper, a new artificial neural network based forward input selection scheme is proposed. The objective of the proposed scheme is to select the smallest number of important variables as model inputs, which will then be used for neural-fuzzy...
An iterative bootstrapping-based data over-sampling strategy is presented in this paper together with an adaptive neural-fuzzy inference system (ANFIS) to deal with a severely imbalanced data modelling problem. As real industrial data are often very large, containing hundreds of process variables and a huge number of data records, the selection of a compact set of input variables becomes critical...
In this paper, I propose a novel hybrid approach of license plate recognition system based on Neural Network and Image Correlation for classification of characters. I used image processing for segmentation. The purpose of this study is to develop a more reliable hybrid system than individual one. The license plate number of the vehicles taken from an acceptable distance from it up to 10m. This hybrid...
Utilizing the artificial neural networks and grey set pare analysis(GSPA), this paper presents a model forecasting the infection rate of computer viruses based on the percentage of four major consequences of virus infection: browser hijack, account theft, illegal remote control as well as system or network failure. The correlation between the infection rate of computer viruses and four other factors...
Support vector machine (SVM) is employed to quantitatively predict the toxicity of 65 aromatic compounds with diverse structure features, and the obtained results are compared systematically with multiple linear regression (MLR), partial least square regression (PLS) and artificial neural network (ANN). It is suggested that SVM possesses higher modeling stability and better generalization ability,...
Gaussian regularization is an effective method to improve the generalization ability of neural networks. A Gaussian regularization RBF neural network (GRNN) which combines the advantages of RAN, and regularization is proposed in this paper. And a model using GRNN is presented to predict the ash fusion temperature (AFT) for some Chinese coals Compared with the traditional techniques, the GRNN prediction...
This paper presents the investigations of forecasting performance of different type of Feedforward Neural Networks (FNN) in forecasting the sunspot numbers. Feedforward Neural Network will be used in this investigation by using different learning algorithms, sunspot data models and FNN transfer functions. Simulations are done using Matlab 7 where customized Graphic User Interface (GUI) called ‘Sunspot...
Segmentation of speech signals is a crucial task in many types of speech analysis. We present a novel approach at segmentation on a syllable level, using a Bidirectional Long-Short-Term Memory Neural Network. It performs estimation of syllable nucleus positions based on regression of perceptually motivated input features to a smooth target function. Peak selection is performed to attain valid nuclei...
Stereo audio enhancement and upmixing techniques require spatial analysis of the mixture in order to work optimally for different types of contents. In this paper a method is proposed which classifies the time-frequency regions in stereo audio data into six different classes. The individual classes represent special cases of a generic stereo signal model which is introduced and characterized in the...
Multi-class object recognition is a critical capability for an intelligence robot to perceive its environment. In this paper, a new approach consisting of a number of modular neural networks is proposed to recognize multiple classes of objects for a robotic system. The population of the modular neural networks depends on the class number of the objects to be recognized and each modular network only...
This paper proposes a faint signal processing approach combining AR model and BP neural network (NN), by which the faint signal is fitted with AR model, whose coefficient served as signal eigenvector, and then sent into a three-tier BP NN for training and recognition classification. Classification tests on human pulse signals between drug users and non-users show that this approach is characterized...
Forecasting short-term electricity market prices has been the focus of several studies in recent years. Although various approaches have been examined, achieving sufficiently low forecasting errors has not been always possible. However, certain applications, such as demand-side management, do not require exact values for future prices but utilize averages values as the basis for making short-term...
A new algorithm is herein developed for combining the classification decisions of different feedforward neural network models with applications to face detection in complex backgrounds. Instead of the usual approach for applying voting schemes or linear combinations schemes on the decisions of their output layer neurons, the proposed methodology integrates higher order representation patterns extracted...
Long range dependence is closely linked with self-similar stochastic processes and random fractals, which have been considered extensively for signal processing applications and computer network traffic modeling. The Hurst parameter captures the amount of long-range dependence in a time series. Typically, the analysis of self-similar series is performed using: the variance-time plot, the R/S plot,...
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.