Performance of small-world feedforward neural networks for the diagnosis of diabetes

Erkaymaz, Okan | Özer, Mahmut | Perc, Matjaž

We investigate the performance of two different small-world feedforward neural networks for the diagnosis of diabetes. We use the Pima Indians Diabetic Dataset as input. We have previously shown than the Watts–Strogatz small-world feedforward neural network delivers a better classification performance than conventional feedforward neural networks. Here, we compare this performance further with the one delivered by the Newman–Watts small-world feedforward neural network, and we show that the latter is better still. Moreover, we show that Newman–Watts small-world feedforward neural networks yiel ...Daha fazlası

Improving genetic algorithms' performance by local search for continuous function optimization

Hamzaçebi C.

The genetic algorithms (GAs) can be used as a global optimization tool for continuous and discrete functions problems. However, a simple GA may suffer from slow convergence, and instability of results. GAs' problem solution power can be increased by local searching. In this study a new local random search algorithm based on GAs is suggested in order to reach a quick and closer result to the optimum solution. © 2007 Elsevier Inc. All rights reserved.

Integrating GPS with national networks by collocation method

Kutoglu H.S. | Ayan T. | Mekik Ç.

In the integration of geodetic networks with different geodetic coordinate systems, the similarity transformation seems to be the most suitable method because the geodetic coordinate systems are designated to be Cartesian (orthogonal). However, this method may always not be practical to use in most surveying works demanding such high accuracies as the coordinates used in the transformation are distorted. In this case, methods alternative to well-known similarity transformations can be considered for the integration of geodetic networks with other systems. However, in this paper, the collocatio ...Daha fazlası

Estimation of the beach bar parameters using the genetic algorithms

Kömürcü M.I. | Tutkun N. | Özölçer I.H. | Akpinar A.

Waves, topographic characteristics and material properties are the most significant factors, which affect the sediment movement and coastal profiles. In this study, considering the wave height (H0) and period (T), the bed slope (m) and the sediment diameter (d50), the cross-shore sediment movement is investigated using a physical model and obtained 80 experimental data for offshore bar geometric parameters. The experimental results are also evaluated by the genetic algorithms (GAs) that are limitedly employed in coastal engineering applications. The results of GAs model and equations cited in ...Daha fazlası

Influences of autapse and channel blockage on multiple coherence resonance in a single neuron

Uzun R.

We study how the spiking regularity of a single stochastic Hodgkin–Huxley neuron is effected in the presence of ion channel blocking and autaptic connection. In this study, we consider a chemical autapse expressed by its coupling strength and delay time. It is found that the neuron exhibits multiple coherence resonance (MCR) behavior induced by autaptic time delay at an appropriate level of ion channel blocking and autaptic coupling strength. This MCR behavior increases with the decrement of working potassium ion channels, whereas it decreases or completely disappears with the increment of a f ...Daha fazlası

The role of common point distribution in obtaining reliable parameters for coordinate transformation

Kutoglu H.S. | Ayan T.

A datum transformation between two geodetic datums primarily requires to compute transformation (datum) parameters through the common points whose the coordinates are available in both datums. The coordinates are burdened with errors caused by different reasons. During the estimation, some parts of these errors are absorbed into the parameters. Therefore, different combinations of common points lead to different sets of the datum parameters. Amounts of the differences between the different parameter sets are dependent on the amounts of the errors which are absorbed in the estimation. In this s ...Daha fazlası

A heuristic approach for finding the global minimum: Adaptive random search technique

Hamzaçebi C. | Kutay F.

In this paper, a new random search technique which facilitates the determination of the global minimum, is presented. This method, called Adaptive Random Search Technique (ARSET), is experimented on test problems, and successful results are obtained. ARSET algorithm, outcome of which is observed to be relatively better, is also compared with other methods. In addition, applicability of the algorithm on artificial neural network training is tested with XOR problem. © 2005 Elsevier Inc. All rights reserved.

A mixed algorithm for numerical computation of soliton solutions of the coupled KdV equation: Finite difference method and differential quadrature method

Başhan A.

The aim of the manuscript is to investigate numerical solutions of the system of coupled Korteweg-de Vries equation. For this approximation, we have used finite difference method for time integration and differential quadrature method depending on modified cubic B-splines for space integration. To display the accuracy of the present mixed method three famous test problems namely single soliton, interaction of two solitons and birth of solitons are solved and the error norms L2 and L? are computed and compared with earlier works. Comparison of error norms show that present mixed method obtained ...Daha fazlası

Chaotic Signal Induced Delay Decay in Hodgkin-Huxley Neuron

BAYSAL VELİ, YILMAZ ERGİN

The Role of Common Point Distribution in Obtaining Reliable Parameters for Coordinate Transformation

KUTOĞLU ŞENOL HAKAN,tevfik ayan

Estimation of the Beach Bar Parameters Using the Genetic Algorithms

KÖMÜRCÜ MURAT İHSAN,TUTKUN NEDİM,ÖZÖLÇER İSMAİL HAKKI,AKPINAR ADEM