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.
Based on multiple input single output of adaptive fuzzy neural network, this paper design the integrated adaptive fuzzy neural network based on the Takagi-Sugeno type fuzzy rules, adopt a hybrid learning algorithm to train the network connection weights, optimize membership function. Simulation results verified the effectiveness and feasibility of this method.
Hardware implementation of Neural Networks (NNs) provides advantages such as parallelism and real-time capabilities, whereas Probabilistic Neural Networks (PNNs) achieve high accuracy in pattern discrimination. In this paper, a FPGA implementation of a PNN sorting algorithm is proposed to sort spikes. Both Matlab-based and FPGA-based sorting algorithms using a PNN were implemented and evaluated, and...
Based on summing up forecasting method and theory about water-inrush from floor of working face, combined with a lot of practical information analysis, the main five factors controlling water-inrush from floor are found out, which are water yield property of aquifer, water pressure, effective confining stratum thickness of mining floor, fault or broken zone and mining pressure. According to the typical...
Process neural networks (PNN) can only receive time-varying continuous functions, can not receive discrete samples. To solve this problem, a training algorithm of PNN based on piecewise linear interpolation function is proposed. First the discrete data of both sample functions and weight functions are transformed to piecewise linear functions, and then the integrals of product of two linear functions...
Credit scoring and behavioral scoring have become very important credit risk management tasks during the past few years due to the impact of several financial crises. The objective of the proposed study is to explore the performance of behavioral scoring using three commonly discussed data mining techniques-linear discriminant analysis (LDA), backpropagation neural networks (BPN), and support vector...
To obtain all feasible machining processes and their quantitative selection priorities, an intelligent making decision approach combining back-propagation neural network and backward planning is proposed. Uniform design method, which is adapted for the problem of multiple factors and multiple levels, is adopted to build representative sample sets for the neural network. The neural network is trained...
Williams Creativity Test B (WCTB) and Adolescent Scientific Creativity Scale (ASCS) were used to measure the creative affective and scientific creativity for 550 middle school students. Generalized regression neural network (GRNN) and multivariable linear regression (MLR) were used for modeling and testing. The result showed the fitting error of GRNN model was lower than the error of MLR. In the rough...
To obtain all feasible machining methods and their quantitative selection priority, an intelligent making decision approach using back-propagation neural network is proposed. Uniform design method, which is adapted for the problem of multiple factors and multiple levels, is adopted to build representative sample sets for the network. The neural network is trained by an improved back-propagation algorithm...
Based on the artificial neural networks and grey correlation analyze, this paper presents a model forecasting the infection rate of computer viruses according to the number of vulnerabilities, the percentage of viruses infecting via web browsing and downloading and the percentage of viruses infecting via portable storage media. The prediction is realized precisely by MATLAB. The three factors are...
Groundwater table often shows complex nonlinear characteristic. Back Propagation (BP) neural network is increasingly used to predict groundwater table. But man-made selecting the structure of BP neural network has blindness and expends much time. In order to overcome shortcomings of traditional BP neural network, Particle Swarm Optimization (PSO) algorithm was adopted to automatically search BP neural...
The concept of ensemble feature selection has been raised by Optiz in his earlier work. And yet, for models like neural networks, new models should be trained and created for every change in its feature subspace, this problem may become tricky when evolutionary algorithms are used to select features, for the slow-training process of neural networks may dramatically extend the whole process of ensemble...
Significance of equipment fault diagnosis is mainly reflected in lower failure rate; lower maintenance costs reduce maintenance time, increase operating time. Wavelet network is the perfect combination of the theory of wavelet analysis and the theory artificial neural network; it is compatible with the superiority of the wavelet and neural networks. In this paper, the wavelet neural network based...
Aims at the complex and dynamic nature of traffic flow in mountain expressway tunnel, through the analysis of change characteristics of traffic flow, based on BP network improve the existing expressway traffic flow model, this thesis puts forward the Elman dynamic neural network model of traffic flow predicting in mountain expressway tunnel. In practice, this model has the strong operational, we adopt...
We design and implement a system to reduce the risk of heat stress, a recognized occupational health hazard (OHH), in two labor intensive industries using a job-combination approach. A novel feature of the system is employing artificial neural networks (ANNs) as model free estimators to evaluate perceived discomforts (PDs) of workers for different job combinations proposed in the work.
Using guided wave characteristics of dispersion and multi-mode, an inverse method based on neural network (NN) is presented to determine the material properties of Functionally Graded Materials (FGM) plates. The group velocities of several lowest modes at several lower frequencies are used as the inputs of the NN model; the outputs of the NN are the distribution function of the volume fraction of...
Standard BP neural network is a most representative algorithm in the neural network model. But shortcomings exist in its process of application. For example: it's hard to reach global optima, but can easily form local minimum, Low study efficiency and slow convergence rate appear because of the excessive training, the selection of the hidden layer nodes lack of theoretical guidance, in training, there...
This paper presents a technique, called “nearly equivalent neural network (NN) model” in the application of nonlinear system identification. This technique is expected to adequately to catch the behavior of the nonlinear system. To demonstrate the new technique proposed, the evaporation system of TP decoration film was analyzed. The complex relationship between the film's transmittance...
In this paper we present many studies and comparisons of different methods of vocabulary selection applied to Topic Detection of some Arabic documents. The topic detection is realized by two methods: neuronal network and support vector machines (SVM). We tested and compared different vocabulary selection methods: word frequency per topic, entropy, Gini and “fselect” based on SVM.
Recently the Runge-Kutta neural network (RKNN) in series-parallel configuration for identification of ordinary differential equation (ODE) was introduced. The neural network is constructed according to the Runge-Kutta approximation method whereby a precise estimate of the changing rates of the system states is possible. In this contribution we extend the approach of to a general state space representation...
This paper presents a technique in how to searching the global minimum for the supervised neural network training. This technique is developed based on the idea of nearly equivalent model. To demonstrate the new technique proposed, two signal processing studies, including signal recognition and signal modeling were simulated. For a comparison, the same simulations were also performed by using 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.