DSpace Arşivi :: by Yazar "Tezel, Gulay" değerine göre listeleniyor

Yazar "Tezel, Gulay" seçeneğine göre listele

Listeleniyor 1 - 19 / 19

Artificial algae algorithm (AAA) for nonlinear global optimization
(ELSEVIER SCIENCE BV, 2015) Uymaz, Sait Ali; Tezel, Gulay; Yel, Esra
In this study, a novel bio-inspired metaheuristic optimization algorithm called artificial algae algorithm (AAA) inspired by the living behaviors of microalgae, photosynthetic species, is introduced. The algorithm is based on evolutionary process, adaptation process and the movement of microalgae. The performance of the algorithm has been verified on various benchmark functions and a real-world design optimization problem. The CEC'05 function set was employed as benchmark functions and the test results were compared with the algorithms of Artificial Bee Colony (ABC), Bee Algorithm (BA), Differential Evolution (DE), Ant Colony Optimization for continuous domain (ACO(R)) and Harmony Search (HSpop). The pressure vessel design optimization problem, which is one of the widely used optimization problems, was used as a sample real-world design optimization problem to test the algorithm. In order to compare the results on the mentioned problem, the methods including ABC and Standard PSO (SPS02011) were used. Mean, best, standard deviation values and convergence curves were employed for the analyses of performance. Furthermore, mean square error (MSE), root mean square error (RMSE) and mean absolute percentage error (MAPE), which are computed as a result of using the errors of algorithms on functions, were used for the general performance comparison. AAA produced successful and balanced results over different dimensions of the benchmark functions. It is a consistent algorithm having balanced search qualifications. Because of the contribution of adaptation and evolutionary process, semi-random selection employed while choosing the source of light in order to avoid local minima, and balancing of helical movement methods each other. Moreover, in tested real-world application AAA produced consistent results and it is a stable algorithm. (C) 2015 Elsevier B.V. All rights reserved.
Artificial algae algorithm with multi-light source for numerical optimization and applications
(ELSEVIER SCI LTD, 2015) Uymaz, Sait Ali; Tezel, Gulay; Yel, Esra
Artificial algae algorithm (AAA), which is one of the recently developed bio-inspired optimization algorithms, has been introduced by inspiration from living behaviors of microalgae. In AAA, the modification of the algal colonies, i.e. exploration and exploitation is provided with a helical movement. In this study, AAA was modified by implementing multi-light source movement and artificial algae algorithm with multi-light source (AAA(ML)) version was established. In this new version, we propose the selection of a different light source for each dimension that is modified with the helical movement for stronger balance between exploration and exploitation. These light sources have been selected by tournament method and each light source are different from each other. This gives different solutions in the search space. The best of these three light sources provides orientation to the better region of search space. Furthermore, the diversity in the source space is obtained with the worst light source. In addition, the other light source improves the balance. To indicate the performance of AAA with new proposed operators (AAA(ML)) experiments were performed on two different sets. Firstly, the performance of AAA and AAA(ML) was evaluated on the IEEE-CEC'13 benchmark set. The second set was real-world optimization problems used in the IEEE-CEC'11. To verify the effectiveness and efficiency of the proposed algorithm, the results were compared with other state-of-the-art hybrid and modified algorithms. Experimental results showed that the multi-light source movement (MLS) increases the success of the AAA. (C) 2015 Elsevier Ireland Ltd. All rights reserved.
Automatic detection of sleep spindles with the use of STFT, EMD and DWT methods
(SPRINGER LONDON LTD, 2018) Yucelbas, Cuneyt; Yucelbas, Sule; Ozsen, Seral; Tezel, Gulay; Kuccukturk, Serkan; Yosunkaya, Sebnem
Sleep staging is a significant process to diagnose sleep disorders. Like other stages, several parameters are required for the determination of N-REM2 stage. Sleep spindles (SSs) are among them. In this study, a methodology was presented to automatically determine starting and ending positions of SSs. To accomplish this, short-time Fourier transform-artificial neural networks (STFT-ANN), empirical mode decomposition (EMD) and discrete wavelet transform (DWT) methods were used. Two considerable methods which were determination envelope and thresholding of the decomposed signals by EMD and DWT were also presented in this study. A database from the EEG signals of nine healthy subjects-which consisted of 100 epochs including 172 SSs in total-was prepared. According to the test results, the highest sensitivity rate was obtained as 100 and 99.42 % for EMD and DWT methods. However, the sensitivity rate for the STFT-ANN method was recorded as 55.93 %. The results indicated that the EMD method could be confidently used in the automatic determination of SSs. Thanks to this study, the sleep experts will be able to reliably find out the epochs with SSs and also know the places of SSs in these epochs, automatically. Another important point of the study was that the sleep staging process-tiring, time-consuming and high fallibility for the experts-could be performed in less time and at higher accuracy rates.
Detection of the Electrode Disconnection in Sleep Signals
(IEEE, 2015) Yucelbas, Cuneyt; Ozsen, Seral; Yucelbas, Sule; Tezel, Gulay; Dursun, Mehmet; Yosunkaya, Sebnem; Kuccukturk, Serkan
Sleep staging process that is performed in sleep laboratories in hospitals has an important role in diagnosing some of the sleep disorders and disturbances which are seen in sleep. And also it is an indispensable method. It is usually performed by a sleep expert through examining during the night of the patients (6-8 hours) recorded Electroencephalogram (EEG), Electrooculogram (FOG), Electromyogram (EMG), electrocardiogram (ECG) and other some signals of the patients and determining the stages of sleep in different time sections named as epochs. Manual sleep staging is preferred among the sleep experts but because it is rather tiring and time consuming task, automatic sleep stage scoring studies has come to the fore. However, none of the so far made automatic sleep staging was not accepted by the experts. The most important reason is that the results of the automated systems are not desired accuracy. There are many factors that affecting the accuracy of the systems, such as noise, the inter-channel interference, excessive body movements and disconnection of electrodes. In this study, we examined the written an algorithm to be able to determine to what extent the disconnection of electrodes in EEG signal that obtained one healthy person at the sleep laboratory of Meram Medicine Faculty of Necmettin Erbakan University. According to the obtained application results, the electrodes disconnection in EEG signal could be detected maximum of 100% and minimum of 99.12% accuracy. Accordingly, based on the success achieved in the study, this algorithm is thought to contribute positively to the researchers that the work on and will work on sleep staging problems and increase the success of automatic sleep staging systems.
Effect of the Hilbert-Huang Transform Method on Sleep Staging
(IEEE, 2017) Yucelbas, Cuneyt; Yucelbas, Sule; Ozsen, Seral; Tezel, Gulay; Yosunkaya, Sebnem
Sleep scoring is performed by examining the recorded electroencephalogram (EEC) and some other signals recorded by a polysomnography (PSG) device. This process is considered more reliable as it is done manually by experts. However, due to the fact that experts may also be mistaken, it has led to an increase in the importance given to automatic sleep staging studies. Many methods have been tested on the signals in order to increase the success of these systems. In this study, an automatic sleep staging system was implemented using the Hilbert-Huang transformation method which is a new time-frequency analysis type. In the study, EEG signals from 5 subjects were used in the sleep laboratory. In the 5-class (Alpha, Beta, Theta, Delta and Spindle bands) applications, the highest classification success was 84.75% as a result of sequential feature selection method.
Elimination of EMG Artifacts From EEG Signal in Sleep Staging
(IEEE, 2016) Ozsen, Seral; Yucelbas, Cuneyt; Yucelbas, Sule; Tezel, Gulay; Yosunkaya, Sebnem; Kuccukturk, Serkan
Sleep staging is a tiring and time-consuming process for the experts. Thus, attention given to automatic sleep staging studies is increasing gradually. Many factors such as effects of EOG and EKG signals to EEG result in contaminated signals rather than clear recorded signals. EMG contamination to EEG is among that kind of factors. In this study, some filters and Discrete Wavelet Transform based EMG artifact elimination process were evaluated on the performance of sleep staging process. Features were extracted from cleaned EEG signals and subjected to a classifier to conduct sleep staging. By using test classification accuracy as a measure of performance, the method giving highest accuracy was tried to be found. Artificial Neural Networks was used in the applications and Discrete Wavelet Transform was found to be the method giving the highest accuracy.
Epilepsy diagnosis using artificial neural network learned by PSO
(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEY, 2015) Yalcin, Nesibe; Tezel, Gulay; Karakuzu, Cihan
Electroencephalogram (EEG) is used routinely for diagnosis of diseases occurring in the brain. It is a very useful clinical tool in the classification of epileptic seizures and the diagnosis of epilepsy. In this study, epilepsy diagnosis has been investigated using EEG records. For this purpose, an artificial neural network (ANN), widely used and known as an active classification technique, is applied. The particle swarm optimization (PSO) method, which does not need gradient calculation, derivative information, or any solution of differential equations, is preferred as the training algorithm for the ANN. A PSO-based neural network (PSONN) model is diversified according to PSO versions, and 7 PSO-based neural network models are described. Among these models, PSONN3 and PSONN4 are determined to be appropriate models for epilepsy diagnosis due to having better classification accuracy. The training methods-based PSO versions are compared with the backpropagation algorithm, which is a traditional method. In addition, different numbers of neurons, iterations/generations, and swarm sizes have been considered and tried. Results obtained from the models are evaluated, interpreted, and compared with the results of earlier works done with the same dataset in the literature.
Epilepsy Diagnosis Using Artificial Neural Network Learned by PSO -- 2
(TURGUT OZAL UNIV, 2012) Yalcin, Nesibe; Tezel, Gulay; Karakuzu, Cihan
In this paper, epilepsy diagnosis has been investigated by using Electroencephalogram (EEG) records. For this purpose, a technique as the classifier Artificial Neural Networks (ANN), which is frequently used and known as an active classification technique, is used. Particle Swarm Optimization (PSO) method is preferred as training algorithm for ANN. PSO based neural network model (PSONN) is diversified according to PSO variants and seven PSO based neural network models are described. In these models, PSONN3 and PSONN4 are determined as appropriate models for the classification. In addition, different number of neurons, iterations/generations and swarm sizes have been considered and tried. Obtained results of the models have been evaluated.
Epilepsy Diagnosis Using PSO based ANN
(ALIFE ROBOTICS CO, LTD, 2013) Yalcin, Nesibe; Karakuzu, Cihan; Tezel, Gulay
Electroencephalogram (EEG) is used routinely for diagnosis of diseases occurring in the brain. It is a very useful clinical tool in classification of epileptic attacks and epilepsy diagnosis. In this paper, epilepsy diagnosis by evaluation of EEG records is presented. Artificial Neural Networks (ANN) is used as a classification technique. Particle Swarm Optimization (PSO) method, which doesn't require gradient calculation, derivative information and any solution of differential equations is preferred for ANN training. This training method is compared with back propagation algorithm, which is one of the traditional methods, and the results are interpreted. In case of using the PSO algorithm, the training and test classification accuracies are %99.67 and %100, respectively. PSO based neural network model (PSONN) has a better classification accuracy than back propagation neural network model (BPNN) for epilepsy diagnosis.
Estimation of the Change in Lake Water Level by Artificial Intelligence Methods
(SPRINGER, 2014) Buyukyildiz, Meral; Tezel, Gulay; Yilmaz, Volkan
In this study, five different artificial intelligence methods, including Artificial Neural Networks based on Particle Swarm Optimization (PSO-ANN), Support Vector Regression (SVR), Multi- Layer Artificial Neural Networks (MLP), Radial Basis Neural Networks (RBNN) and Adaptive Network Based Fuzzy Inference System (ANFIS), were used to estimate monthly water level change in Lake Beysehir. By using different input combinations consisting of monthly Inflow - Lost flow (I), Precipitation (P), Evaporation (E) and Outflow (O), efforts were made to estimate the change in water level (L). Performance of models established was evaluated using root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R-2). According to the results of models, epsilon-SVR model was obtained as the most successful model to estimate monthly water level of Lake Beysehir.
A genetic algorithm-support vector machine method with parameter optimization for selecting the tag SNPs
(ACADEMIC PRESS INC ELSEVIER SCIENCE, 2013) Ilhan, Ilhan; Tezel, Gulay
SNPs (Single Nucleotide Polymorphisms) include millions of changes in human genome, and therefore, are promising tools for disease-gene association studies. However, this kind of studies is constrained by the high expense of genotyping millions of SNPs. For this reason, it is required to obtain a suitable subset of SNPs to accurately represent the rest of SNPs. For this purpose, many methods have been developed to select a convenient subset of tag SNPs, but all of them only provide low prediction accuracy. In the present study, a brand new method is developed and introduced as GA-SVM with parameter optimization. This method benefits from support vector machine (SVM) and genetic algorithm (GA) to predict SNPs and to select tag SNPs, respectively. Furthermore, it also uses particle swarm optimization (PSO) algorithm to optimize C and gamma parameters of support vector machine. It is experimentally tested on a wide range of datasets, and the obtained results demonstrate that this method can provide better prediction accuracy in identifying tag SNPs compared to other methods at present. (c) 2012 Elsevier Inc. All rights reserved.
How to Select Tag SNPs in Genetic Association Studies? The CLONTagger Method with Parameter Optimization
(MARY ANN LIEBERT, INC, 2013) Ilhan, Ilhan; Tezel, Gulay
Selection of genetic variants is a crucial first step in the rational design of studies aimed at explaining individual differences in susceptibility to complex human diseases or health intervention outcomes; for example, in the emerging fields of pharmacogenomics, nutrigenomics, and vaccinomics. While single nucleotide polymorphisms (SNPs) are frequently employed in these studies, the cost of genotyping a huge number of SNPs remains a limiting factor, particularly in low and middle income countries. Therefore, it is important to detect a subset of SNPs to represent the rest of SNPs with maximum possible accuracy. The present study introduces a new method, CLONTagger with parameter optimization, which uses Support Vector Machine (SVM) to predict the rest of SNPs and Clonal Selection Algorithm (CLONALG) to select tag SNPs. Furthermore, the Particle Swarm Optimization algorithm is preferred for the optimization of C and gamma parameters of the Support Vector Machine. Additionally, using many datasets, we compared the proposed new method with the tag SNP selection algorithms present in literature. Our results suggest that the CLONTagger with parameter optimization can identify tag SNPs with better prediction accuracy than other methods. Application-oriented studies are warranted to evaluate the utility of this method in future research in human genetics and study of the genetic components of variable responses to drugs, nutrition, and vaccines.
Monthly evaporation forecasting using artificial neural networks and support vector machines
(SPRINGER WIEN, 2016) Tezel, Gulay; Buyukyildiz, Meral
Evaporation is one of the most important components of the hydrological cycle, but is relatively difficult to estimate, due to its complexity, as it can be influenced by numerous factors. Estimation of evaporation is important for the design of reservoirs, especially in arid and semi-arid areas. Artificial neural network methods and support vector machines (SVM) are frequently utilized to estimate evaporation and other hydrological variables. In this study, usability of artificial neural networks (ANNs) (multilayer perceptron (MLP) and radial basis function network (RBFN)) and epsilon-support vector regression (SVR) artificial intelligence methods was investigated to estimate monthly pan evaporation. For this aim, temperature, relative humidity, wind speed, and precipitation data for the period 1972 to 2005 from Beysehir meteorology station were used as input variables while pan evaporation values were used as output. The Romanenko and Meyer method was also considered for the comparison. The results were compared with observed class A pan evaporation data. In MLP method, four different training algorithms, gradient descent with momentum and adaptive learning rule backpropagation (GDX), Levenberg-Marquardt (LVM), scaled conjugate gradient (SCG), and resilient backpropagation (RBP), were used. Also, epsilon-SVR model was used as SVR model. The models were designed via 10-fold cross-validation (CV); algorithm performance was assessed via mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R (2)). According to the performance criteria, the ANN algorithms and epsilon-SVR had similar results. The ANNs and epsilon-SVR methods were found to perform better than the Romanenko and Meyer methods. Consequently, the best performance using the test data was obtained using SCG(4,2,2,1) with R (2) = 0.905.
A new approach to eliminating EOG artifacts from the sleep EEG signals for the automatic sleep stage classification
(SPRINGER, 2017) Dursun, Mehmet; Ozsen, Seral; Yucelbas, Cuneyt; Yucelbas, Sule; Tezel, Gulay; Kuccukturk, Serkan; Yosunkaya, Sebnem
Interference between EEG and EOG signals has been studied heavily in clinical EEG signal processing applications. But, in automatic sleep stage classification studies these effects are generally ignored. Thus, the objective of this study was to eliminate EOG artifacts from the EEG signals and to see the effects of this process. We proposed a new scheme in which EOG artifacts are separated from electrode or other line artifacts by a correlation and discrete wavelet transform-based rule. Also, to discriminate the situation of EEG contamination to EOG from EOG contamination to EEG, we introduced another rule and integrated this rule to our proposed method. The proposed method was also evaluated under two different circumstances: EOG-EEG elimination along the whole 0.3-35 Hz power spectrum and EOG-EEG elimination with discrete wavelet transform in 0-4 Hz frequency range. To see the consequences of EOG-EEG elimination in these circumstances, we classified pure EEG and artifact-eliminated EEG signals for each situation with artificial neural networks. The results on 11 subjects showed that pure EEG signals gave a mean classification accuracy of 60.12 %. The proposed EOG elimination process performed in 0-35 Hz frequency range resulted in a classification accuracy of 63.75 %. Furthermore, conducting EOG elimination process by using 0-4 Hz DWT detail coefficients caused this accuracy to be raised to 68.15 %. By comparing the results obtained from all applications, we concluded that an improvement about 8.03 % in classification accuracy with regard to the uncleaned EEG signals was achieved.
A novel system for automatic detection of K-complexes in sleep EEG
(SPRINGER LONDON LTD, 2018) Yucelbas, Cuneyt; Yucelbas, Sule; Ozsen, Seral; Tezel, Gulay; Kuccukturk, Serkan; Yosunkaya, Sebnem
Sleep staging process is applied to diagnose sleep-related disorders by sleep experts through analyzing sleep signals such as electroencephalogram (EEG), electrooculogram and electromyogram of subjects and determining the stages in 30-s-length time parts of sleep named as epochs. Subjects enter several stages during the sleep, and N-REM2 is one of them which has also the highest duration among the other stages. Approximately half of the sleep consists of N-REM2. One of the important parameters in determining N-REM2 stage is K-complex (Kc). In this study, some time and frequency analysis methods were used to determine the locations of Kcs, automatically. These are singular value decomposition (SVD), variational mode decomposition and discrete wavelet transform. The performance of them in detecting Kcs was compared. Furthermore, systems with combinations of these methods were presented with logic AND operations. The EEG recordings of seven subjects were obtained from the Sleep Research Laboratory of Necmettin Erbakan University. A database with total 359 Kcs in 320 epochs was prepared from the recordings. According to the results, the highest average recognition rate was found as 92.29% for the SVD method. Thanks to this study, the sleep experts can find out whether there were Kcs in related epochs and also know their locations in these epochs, automatically. Also, it will help automatic sleep stage classification systems.
Pre-determination of OSA degree using morphological features of the ECG signal
(PERGAMON-ELSEVIER SCIENCE LTD, 2017) Yucelbas, Sule; Yucelbas, Cuneyt; Tezel, Gulay; Ozsen, Seral; Kuccukturk, Serkan; Yosunkaya, Sebnem
Obstructive sleep apnea (OSA) is a very common, but a difficult sleep disorder to diagnose. Recurrent obstructions form in the airway during sleep, such that OSA can threaten a breathing capacity of patients. Clinically, continuous positive airway pressure (CPAP) is the most specific and effective treatment for this. In addition, these patients must be separated according to its degree, with CPAP treatment applied as a result. In this study, 30 OSA patients from two different databases were automatically classified using electrocardiogram (ECG) data, identified as mild, moderate, and severe. One of the databases was original recordings which had 9 OSA patients with 8303 epochs and the other one was Physionet benchmark database which had 21 patients with 20,824 epochs. Fifteen morphological features could be identified when apnea was seen, both before and after it presented. Five data groups in total for first dataset and second dataset were prepared with these features and 10-fold cross validation was used to effectively determine the test data. Then, sequential backward feature selection (SBFS) algorithm was applied to understand the more effective features. The prepared data groups were evaluated with artificial neural networks (ANN) to obtain optimum classification performance. All processes were repeated for ten times and error deviation was calculated for the accuracy. Furthermore, different classifiers which are frequently used in the literature were tested with selected features. The degree of OSA was estimated from three epochs in pre-apnea data, yielding the success rates of 97.20 +/- 2.15% and 90.18 +/- 8.11% with the SBFS algorithm for the first and second datasets, respectively. Also, SVM classifier followed ANN system in the success rates of 96.23 +/- 3.48% and 88.75 +/- 8.52% for used datasets. (C) 2017 Elsevier Ltd. All rights reserved.
Tag SNP selection using clonal selection and majority voting algorithms
(INDERSCIENCE ENTERPRISES LTD, 2016) Ilhan, Ilhan; Tezel, Gulay
Researchers should select a suitable subgroup that includes all SNPs and represents the rest of the SNPs with little error for very large-scale association studies. The SNPs included in the subgroup are tag SNPs or haplotype tag SNPs (htSNPs). When selecting the tag SNPs, it is critical to accurately predict and identify the smallest number of tag SNPs with minimum error. This study used the Clonal Selection Algorithm (CLONALG) to decide on the tag SNPs to be included in the subgroup. In addition, the study proposed a new method called CSMV, which used the Majority Voting (MV) method to predict the rest of the SNPs. This method was compared with the BPSO method and the CLONTagger with parameter optimisation method using datasets of different sizes. According to the experimental results of the study, the CSMV method could determine the tag SNPs with significantly higher accuracy than the other two methods.
Tag SNP Selection Using Similarity Associations Between SNPs
(IEEE, 2015) Ilhan, Ilhan; Tezel, Gulay; Ozcan, Cengiz
Genetic changes that may be associated with complex diseases are tried to be determined by means of many genome-wide association studies. Single Nucleotide Polymorphisms ( SNPs) are used primarily in these studies since they comprise a large part of these genetic changes. Statistical importance of the genome-wide association study is directly related to the number of individuals and SNPs. However, it is still very costly and time-consuming to genotype all SNPs inside the candidate area for many individuals in very large-scale association studies. For this reason, with a small error, it is necessary to select an appropriate subset of all SNPs that will represent the rest of SNPs. These selected SNPs are called tag SNPs or haplotype tag SNPs ( tag SNPs or htSNPs). It is essential in tag SNP selection to determine minimum tag SNP set with very good prediction accuracy. In this study, while Clonal Selection Algorithm ( CLONALG) was used as tag SNP selection method, a new method named CLONSim, in which similarity association between SNPs was used as the prediction method for the rest of SNPs was proposed. The proposed method was compared with BPSO ( Binary Particle Swarm Optimization) and CLONTagger methods with parameter optimization using datasets of different sizes. Experiment results showed that the proposed method could identify tag SNPs significantly faster.
Utilization of PSO algorithm in estimation of water level change of Lake Beysehir
(SPRINGER WIEN, 2017) Buyukyildiz, Meral; Tezel, Gulay
In this study, unlike backpropagation algorithm which gets local best solutions, the usefulness of particle swarm optimization (PSO) algorithm, a population-based optimization technique with a global search feature, inspired by the behavior of bird flocks, in determination of parameters of support vector machines (SVM) and adaptive network-based fuzzy inference system (ANFIS) methods was investigated. For this purpose, the performances of hybrid PSO-epsilon support vector regression (PSO-epsilon SVR) and PSO-ANFIS models were studied to estimate water level change of Lake Beysehir in Turkey. The change in water level was also estimated using generalized regression neural network (GRNN) method, an iterative training procedure. Root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R (2)) were used to compare the obtained results. Efforts were made to estimate water level change (L) using different input combinations of monthly inflow-lost flow (I), precipitation (P), evaporation (E), and outflow (O). According to the obtained results, the other methods except PSO-ANN generally showed significantly similar performances to each other. PSO-epsilon SVR method with the values of minMAE = 0.0052 m, maxMAE = 0.04 m, and medianMAE = 0.0198 m; minRMSE = 0.0070 m, maxRMSE = 0.0518 m, and medianRMSE = 0.0241 m; minR (2) = 0.9169, maxR (2) = 0.9995, medianR (2) = 0.9909 for the I-P-E-O combination in testing period became superior in forecasting water level change of Lake Beysehir than the other methods. PSO-ANN models were the least successful models in all combinations.

Yazar "Tezel, Gulay" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri