Özer, Mahmut | Graham, Lyle J. | Erkaymaz, Okan | Uzuntarla, Muhammet

Article | 2007 | NeuroReport18 ( 13 ) , pp.1371 - 1374

Cortical neurons in-vivo operate in a continuum of overall conductance states, depending on the average level of background synaptic input throughout the dendritic tree. We compare how variability, or fluctuations, in this input affects the statistics of the resulting 'spontaneous' or 'background' firing activity, between two extremes of the mean input corresponding to a low-conductance (LC) and a high-conductance (HC) state. In the HC state, we show that both firing rate and regularity increase with increasing variability. In the LC state, firing rate also increases with input variability, but in contrast to the HC state, firing re . . .gularity first decreases and then increases with an increase in the variability. At high levels of input variability, firing regularity in both states converge to similar values. © 2007 Lippincott Williams & Wilkins, Inc Daha fazlası Daha az

Erkaymaz, Okan | Özer, Mahmut | Perc, Matjaž

Article | 2017 | Applied Mathematics and Computation311 , pp.22 - 28

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 yield the highest output correlation as well . . . as the best output error parameters. © 2017 Elsevier Inc Daha fazlası Daha az

Erkaymaz, Okan | Özer, Mahmut

Article | 2016 | Chaos, Solitons and Fractals83 , pp.178 - 185

Artificial intelligent systems have been widely used for diagnosis of diseases. Due to their importance, new approaches are attempted consistently to increase the performance of these systems. In this study, we introduce a new approach for diagnosis of diabetes based on the Small-World Feed Forward Artificial Neural Network (SW- FFANN). We construct the small-world network by following the Watts-Strogatz approach, and use this architecture for classifying the diabetes, and compare its performance with that of the regular or the conventional FFANN. We show that the classification performance of the SW-FFANN is better than that of the . . . conventional FFANN. The SW-FFANN approach also results in both the highest output correlation and the best output error parameters. We also perform the accuracy analysis and show that SW-FFANN approach exhibits the highest classifier performance. © 2015 Elsevier Ltd. All rights reserved Daha fazlası Daha az

Erkaymaz, Okan | Özer, Mahmut | Yumuşak, Nejat

Article | 2014 | Turkish Journal of Electrical Engineering and Computer Sciences22 ( 3 ) , pp.708 - 718

Since feed-forward artificial neural networks (FFANNs) are the most widely used models to solve real-life problems, many studies have focused on improving their learning performances by changing the network architecture and learning algorithms. On the other hand, recently, small-world network topology has been shown to meet the characteristics of real-life problems. Therefore, in this study, instead of focusing on the performance of the conventional FFANNs, we investigated how real-life problems can be solved by a FFANN with small-world topology. Therefore, we considered 2 real-life problems: estimating the thermal performance of so . . .lar air collectors and predicting the modulus of rupture values of oriented strand boards. We used the FFANN with small-world topology to solve both problems and compared the results with those of a conventional FFANN with zero rewiring. In addition, we investigated whether there was statistically significant difference between the regular FFANN and small-world FFANN model. Our results show that there exists an optimal rewiring number within the small-world topology that warrants the best performance for both problems. © TUBITAK Daha fazlası Daha az

Uzun, Rukiye | Erkaymaz, Okan | Şenyer Yapıcı, İrem

Article | 2018 | Gazi University Journal of Science31 ( 1 ) , pp.100 - 110

The surface electromyography (sEMG) is useful tool to diagnose of knee disorder in clinical environments. It assists in designing the clinical decision support systems based classification. These systems exhibit complex structure because of sEMG data obtained at different postures at this study. In this context, we have researched the classification performance of each posture using artificial neural network (ANN) and logistic regression (LR) models and have showed that the classification success of the model used sitting posture data is higher than other postures (gait and standing). We have promoted this finding by using machine l . . .earning and statistical methods. The results show that the proposed models can classify with over 95% of success, and also the ANN model has higher performance than the LR model. Our ANN model outperforms reported studies in literature. The accuracy results indicate that the models used the only sitting posture data can exhibit successful classification for the knee disorder. Therefore, the usage of complex dataset is prevented for diagnosing knee disorder. © 2018, Gazi University Eti Mahallesi. All rights reserved Daha fazlası Daha az

Ustabaş, Kaya Gülhan | Erkaymaz, Okan | Saraç, Zehra

Article | 2019 | Soft Computing23 ( 18 ) , pp.8827 - 8837

In this paper,a fuzzy interference and an adaptive neuro-fuzzy interference system models have been presented in order to accelerate designing of the digital holographic setup without experiment. The setting parameters of experimental holographic setup, which affect the quality of images obtained from reconstructed holograms, are predicted digitally by proposed models before the recording process. Hence, we reduce the required time for designing of digital holographic setup with optimization process. The adaptive neuro-fuzzy interference system model for the optimization of the digital holographic setup is first attempt in the liter . . .ature. The accuracy of the proposed models is examined by comparing the presented models and actual calculated experimental root-mean-square values. As a result, the accuracy of the adaptive neuro-fuzzy interference system shows the better performance than the fuzzy interference system. Moreover, the design of experimental setup can be occurred numerically in a short time by using adaptive neuro-fuzzy interference system models. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature Daha fazlası Daha az

Kaya, Gülhan Ustabaş | Erkaymaz, Okan | Saraç, Zehra

Article | 2016 | Expert Systems with Applications56 , pp.177 - 185

In this study, the optimization of the digital holography setup is achieved by a using fuzzy logic prediction system. In fact, when this optimization process is experimentally performed, some parameters are changed in the setup. These parameters affect directly the obtained image quality after a reconstruction process, which is determined by normalized root mean square. The aim of this study is to achieve the optimization of digital holographic setup by using both experimental and fuzzy logic prediction systems. Furthermore, the required time during the experimental optimization can be lowered by using a numerical method like the fu . . .zzy logic prediction system. Here, the experimental optimization results and the optimization results obtained by the fuzzy logic prediction system are compared. It is offered that the designed experimental system can be optimized by using an artificial intelligent tool. The applied fuzzy logic prediction model is used the first time for optimization of hologram recording setup. As a result, it is reached a conclusion that the optimization of digital holographic setup can be numerically performed by the fuzzy logic prediction system. Moreover, while digital holographic setup is experimentally designed, the required time for optimization is reduced, as well. © 2016 Elsevier Ltd. All rights reserved Daha fazlası Daha az

Kaya, Ceren | Erkaymaz, Okan | Ayar, Orhan | Özer, Mahmut

Article | 2018 | Chaos, Solitons and Fractals114 , pp.164 - 174

In this study, we introduce two hybrid artificial neural network models with particle swarm optimization algorithm to diagnose diabetic retinopathy based on the Video-Oculography signals. The hybrid models use Discrete Wavelet Transform and Hilbert-Huang Transform separately to extract features from the signals. The classification performance of both models is analyzed comparatively. We show that the model based on Hilbert–Huang Transform exhibits better classification performance than the model based on the Discrete Wavelet Transform. © 2018 Elsevier Ltd