Designing a wavelet neural network (WNN) needs to be done judiciously in attaining the optimal generalization performance. Its prediction competence relies highly on the initial value of translation vectors. However, there is no established solution in determining the appropriate initial value for the translation vectors at this moment. In this paper, we propose a novel enhanced fuzzy c-means clustering algorithm - specifically, the modified point symmetry-based fuzzy c-means (MPSDFCM) algorithm - in initializing the translation vectors of the WNNs. The effectiveness of embedding different activation functions in WNNs will be investigated as well. The categorization effectiveness of the proposed WNNs model was then evaluated in classifying the type 2 diabetics, and was compared with the multilayer perceptrons (MLPs) and radial basis function neural networks (RBFNNs) models. Performance assessment shows that our proposed model outperforms the rest, since a 100% superior classification rate was achieved.
Color quantization is an important process for image processing and various applications. Up to now, many color quantization methods have been proposed. The self-organizing maps (SOM) method is one of the most effective color quantization methods, which gives excellent color quantization results. However, it is slow, so it is not suitable for real-time applications. In this paper, we present a color importance{based SOM color quantization method. The proposed method dynamically adjusts the learning rate and the radius of the neighborhood using color importance. This makes the proposed method faster than the conventional SOM-based color quantization method. We compare the proposed method to 10 well-known color quantization methods to evaluate performance. The methods are compared by measuring mean absolute error (MAE), mean square error (MSE), and processing time. The experimental results show that the proposed method is effective and excellent for color quantization. Not only does the proposed method provide the best results compared to the other methods, but it uses only 67.18% of the processing time of the conventional SOM method.
The additive mixture rules have been extended for calculation of the effective longitudinal elasticity modulus of the composite (Functionally Graded Materials - FGM's) beams with both the polynomial longitudinal variation of the constituent's elasticity modulus. Stiffness matrix of the composite Bernoulli-Euler beam has been established which contains the transfer constants. These transfer constants describe very accurately the polynomial uni-axially variation of the effective longitudinal elasticity modulus, which is calculated using the extended mixture rules.
The mixture rules have been extended for calculation of the effective elasticity modulus for stretching and flexural bending of the layer-wise symmetric composite (FGM's) sandwich beam finite element as well. The polynomial longitudinal and transversally symmetric layer-wise variation of the sandwich beam stiffness has been taken into the account. Elastic behaviour of the sandwich beam will be modelled by the laminate theory. Stiffness matrix of such new sandwich beam element has been established. The nature and quality of the matrix reinforcement interface have not been considered. Four examples have been solved using the extended mixture rules and the new composite (FGM's) beam elements with varying stiffness. The obtained results are evaluated, discussed and compared. and Obsahuje seznam literatury
diagnosis. Moreover various studies can be found in medical journals dedicated to Artificial Neural Networks (ANN). In the presented study, a method was developed to learn and detect benign and malignant tumor types in contrast-enhanced breast magnetic resonance images (MRI). The backpropagation algorithm was taken as the ANN learning algorithm. The algorithm (NEUBREA) was developed in C# programming language by using Fast Artificial Neural Network Library (FANN). Having been diagnosed by radiologists, 7 cases of malignant tumor, 8 cases of benign tumor, and 3 normal cases were used as a training set. The results were tested on 34 cases that had been diagnosed by radiologists. After the comparison of the results, the overall accuracy of algorithm was defined as 92%.
When digital signals are transmitted through frequency selective communication channels, one of the problems that arise is inter-symbol interference (ISI). To compensate corruptions caused by ISI and to find original information being transmitted, an equalization process is performed at the receiver. Since communication channels are time varying and random in nature, adaptive equalizers must be used to learn and subsequently track the time varying characteristics of the channel. Traditional equalizers are based on finding the inverse of the channel and compensating the channel's influence using inverse filter technique. There exists no equalizer for non-invertible channels. Artificial Neural Networks (ANN) can be applied to this for achieving better performance than conventional methods. We have proposed a model of a neural equalizer using MLP (multi layer perceptron), which reduces the mean square error to minimum and eliminates the effects of ISI. Empirically we have found that this neural equalizer is more efficient than conventional adaptive equalizers.
An efficient estimator for the expectation ∫f\dP is constructed, where P is a Gibbs random field, and f is a local statistic, i. e. a functional depending on a finite number of coordinates. The estimator coincides with the empirical estimator under the conditions stated in Greenwood and Wefelmeyer \cite{greenwood_wefelmeyer_1999}, and covers the known special cases, namely the von Mises statistic for the i.i.d. underlying fields and the case of one-dimensional Markov chains.
One of the most challenging problems in the optimal control theory consists of solving the nonsmooth optimal control problems where several discontinuities may be present in the control variable and derivative of the state variable. Recently some extended spectral collocation methods have been introduced for solving such problems, and a matrix of differentiation is usually used to discretize and to approximate the derivative of the state variable in the particular collocation points. In such methods, there is typically no condition for the continuity of the state variable at the switching points. In this article, we propose an efficient hp spectral collocation method for the general form of nonsmooth optimal control problems based on the operational integration matrix. The time interval of the problem is first partitioned into several variable subintervals, and the problem is then discretized by considering the Legendre-Gauss-Lobatto collocation points. Here, the switching points are unknown parameters, and having solved the final discretized problem, we achieve some approximations for the optimal solutions and the switching points. We solve some comparative numerical test problems to support of the performance of the suggested approach.
Time series forecasting, such as stock price prediction, is one of the most important complications in the financial area as data is unsteady and has noisy variables, which are affected by many factors. This study applies a hybrid method of Genetic Algorithm (GA) and Artificial Neural Network (ANN) technique to develop a method for predicting stock price and time series. In the GA method, the output values are further fed to a developed ANN algorithm to fix errors on exact point. The analysis suggests that the GA and ANN can increase the accuracy in fewer iterations. The analysis is conducted on the 200-day main index, as well as on five companies listed on the NASDAQ. By applying the proposed method to the Apple stocks dataset, based on a hybrid model of GA and Back Propagation (BP) algorithms, the proposed method reaches to 99.99% improvement in SSE and 90.66% in time improvement, in comparison to traditional methods. These results show the performances and the speed and the accuracy of the proposed approach.
Artificial Neural Network (ANN) is the primary automated AI system preferred for medical applications. Even though ANN possesses multiple advantages, the convergence of the ANN is not always guaranteed for the practical applications. This often results in the local minima problem and ultimately yields inaccurate results. This convergence problem is common among ANNs and especially in Kohonen neural networks which employ unsupervised training methodology. In this work, an Efficient Kohonen Fuzzy Neural (EKFN) network is proposed to eliminate the iteration dependent nature of the conventional system. The suitability of this hybrid automated system is illustrated in the context of pathology identification in retinal images. This disease identification system includes anatomical structure segmentation from retinal images followed by image classification. The performance measures used are accuracy, sensitivity, specificity, positive predictive value and positive likelihood ratio. Experimental results show promising possibilities for the hybrid systems in terms of performance measures.
This paper presents an efficient learning algorithm that generates radial basis function neural network with few neurons. The neural network adds neurons according to a growth criterion defined by the current output error, the current input's distance to the nearest center, and the root-mean-square output error over a sliding windows, deletes neurons by a pruning strategy based on the error reduction rates, and updates the output-layer weights with a Givens QR decomposition based on the orthogonalized least square algorithm. Simulations on two benchmark problems demonstrate that the algorithm produces smaller networks than RAN, RANEKF, and MRAN, and consumes less training time than RAN, RANEKF, MRAN, and GAP-RBF.