Yazar "Kutucu, H." seçeneğine göre listele
Listeleniyor 1 - 20 / 25
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Analysis of the early posttraumatic period pathophysiology in case of the severe combined thoracic trauma using multivariate logistic regression(CEUR-WS, 2019) Stupnytskyi, M.; Zhukov, V.; Gorbach, T.; Biletskii, O.; Kutucu, H.Severely injured patients are always challenging, even more so when they have suffered critical trauma to the chest. The aim of this study is to create a prognostic tool for outcome prediction for patients with combined thoracic trauma based on the determination of main homeostasis parameters on the 1st and 2nd day after injury. Multivariate logistic regression analysis with forward elimination of the variables was used for modeling the dependence of outcome on clinical and laboratory parameters that reflects main pathophysiological mechanisms developed on the 1st and 2nd day after combined thoracic trauma. 73 Male patients with combined thoracic trauma were included in the study. The results of fitting a logistic regression model show the relationship between mortality and six independent variables: transferrin saturation, percentage of eosinophils, TNF-a concentration, total iron binding capacity, inspiratory fraction of oxygen and albumin concentration. Besides that, forward elimination of the variables into the logistic regression equation helps to recognize relatively independent pathophysiological mechanisms involved to progression of wound dystrophy. The likelihood ratio tests can reflect the contribution degree of each pathogenesis rout responsible for the negative outcomes of the severe combined thoracic trauma. The study contributes to our understanding of interaction between pathophysiological mechanisms that make harmful effects and are involved in the progression of wound dystrophy and compensatory reactions directed on stabilization of vital function disturbances and maintenance of homeostasis during this type of wound dystrophy. Copyright © 2019 for this paper by its authors.Öğe An application of artificial neural networks to assessment of the wind energy potential in Libya(Institute of Electrical and Electronics Engineers Inc., 2017) Kutucu, H.; Almryad, A.We modeled in this paper the variation of wind speed as a renewable energy in Mediterranean Sea of Libya (North of Africa) using an artificial neural network (ANN). We developed multi-layer, feed-forward, back-propagation artificial neural networks for prediction monthly mean wind speed. The monthly mean wind speed data of 25 cities in Libya were monitored during the period of six years from 2010 to 2015. Meteorological (mean temperature, relative humidity and mean sunshine duration) and geographical data (latitude, longitude and altitude) are used as the inputs and the wind speed is used as the output of the ANN. The experimental results show that the correlation coefficients between the predicted and measured wind speeds for training data sets are higher than 0.99. Therefore, the ANN model can be used with high prediction accuracy at locations where wind speed data are not measured. © 2016 IEEE.Öğe An artificial bee colony algorithm for solving the weapon target assignment problem(Association for Computing Machinery, 2017) Durgut, R.; Kutucu, H.; Akleylek, S.In this paper, we deal with the static weapon target assignment (WTA) problem which is a hard combinatorial optimization problem having some industrial applications. The aim of the WTA problem is to find an assignment of weapons to targets with the minimum total survival value of the targets. The WTA problem is known to be NP-complete problem. In this paper, we propose a novel artificial bee algorithm to give an efficient solution to the WTA problem. We test the proposed algorithm with benchmark problem instances and compare it with some other meta-heuristics in the literature. Computational tests show that our algorithm is competitive. © 2017 Association for Computing Machinery.Öğe The band collocation problem: A library of problems and a metaheuristic approach(CEUR-WS, 2016) Kutucu, H.; Gursoy, A.; Kurt, M.; Nuriyev, U.In this paper, we consider the Band Collocation Problem (BCP) which may find an application in telecommunication networks, to design an optimal packing of information flows on different wavelengths into groups for obtaining the highest available cost reduction using wavelength division multiplexing (WDM) technology. We give a review of its mathematical models. The linear and nonlinear models have been implemented in GAMS (the General Algebraic Modeling System) and solved using the CPLEX and KNITRO solvers, respectively. Then, we introduce the BCP Library (BCPLib) including 1296 problem instances with different properties that can be accessed at http://www.izmir.edu.tr/bps. Finally, we improve a simulated annealing (SA) meta-heuristic to solve the BCP. The proposed algorithm is performed using two local search methods for several test instances of the BCPLib and compared with the solutions obtained by a genetic algorithm. Experimental results showed that the proposed algorithm improves the quality of solutions. Copyright © by the paper's authors.Öğe Detection of bone fractures using image processing techniques and artificial neural networks(Institute of Electrical and Electronics Engineers Inc., 2017) Öztürk, Ö.; Kutucu, H.The use of computer technology in medical sciences is spreading with technology. The use of computers especially for imaging has become a third eye for physicians. In orthopedic surgeons, after simple roentgenograms for fracture detection, the use of computerized tomography and magnetic resonance has provided great convenience in the detection of fracture, typing, and therefore the appropriate treatment of the patient. The advancing technology has increased the quality of the images in the x-rayograms, reduced artifacts and enabled digital measurements. In this study, image processing and learning techniques were used to diagnose long bone fractures. The proposed artificial neural network has 89% success rate. © 2017 IEEE.Öğe GPU implementation of quantum secure ABC cryptosystem on CUDA(CEUR-WS, 2021) Akleylek, S.; Koyutürk, R.; Kutucu, H.In this paper, we consider the ABC cryptosystem based on multivariate polynomial systems which is one of the post-quantum cryptosystems. We review the theoretical structure of the ABC cryptosystem and implement it on the GPU by using the NVIDIA CUDA technology. We carry out the GPU and CPU implementation details of the ABC cryptosystem on three computers with different graphics cards. We also give a comprehensive comparison between the implementations. We compute the required number of arithmetic operations for each process: key generation, encryption and decryption. According to the experimental results, the GPU implementations have better memory performance than the CPU implementations. Moreover, the encryption process is faster in the GPU implementation. Due to the structure of ABC cryptosystem, the decryption process is slower in the GPU implementation. © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org)Öğe A heuristic algorithm for the band collocation problem(Institute of Electrical and Electronics Engineers Inc., 2017) Gursoy, A.; Kurt, M.; Kutucu, H.; Nuriyev, U.In this paper we present a heuristic algorithm for the The Band Collocation Problem (BCP) which may have some applications in the field of telecommunication. First, we give the definition the BCP. Second, we explain how we create the problem instances with known optimal solutions as a library. Third, we propose the heuristic algorithm. Then, we analysis and interpret the results of the proposed algorithm on the problem instances with known optimal solutions. Finally, we suggest new ideas about the BCP and its solution approaches. © 2016 IEEE.Öğe Heuristic architecture search using network morphism for Chest X-Ray classification(CEUR-WS, 2020) Radiuk, P.; Kutucu, H.Nowadays, the demand for medical image computing is exceptionally high. This growth was mostly driven by the manual development of machine learning models, in particular neural networks. However, due to the constant evolution of domain requirements, manual model development has become insufficient. The present study proposes a heuristic architecture search that can be in an excellent service for the task of medical image classification. We implemented a novel approach called network morphism to the search algorithm. The proposed search method utilizes the enforced hill-climbing algorithm and functional-saving modifications. As a result of computational experiments, the search method found the optimal architecture in 28 GPU hours. The model formed by the found architecture achieved performance of 73.2% in validation accuracy and 84.5% in AUC on the validation dataset that is competitive to the state-of-the-art hand-crafted networks. Moreover, the proposed search method managed to find the architecture that contains four times fewer parameters. Besides, the model requires almost ten times less physical memory, which may indicate the practical usefulness of our method in medical image analysis. Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). IntelITSIS-2020Öğe Hybridization of the SGTM Neural-Like Structure Through Inputs Polynomial Extension(Institute of Electrical and Electronics Engineers Inc., 2018) Vitynskyi, P.; Tkachenko, R.; Izonin, I.; Kutucu, H.In this paper, a new approach for increasing the approximation accuracy with the use of computational intelligence tools is described. It is based on the compatible use of the neural-like structure of the Successive Geometric Transformations Model and the inputs polynomial extension. To implement such an extension, second degree Wiener polynomial is used. This combination improves the method accuracy for solving various tasks, such as classification and regression, including short-term and long-term prediction, dynamic pricing, as well as image recognition and image scaling, e-commerce. Due to the use of SGTM neural-like structure, the high speed of the system is maintained in both training and using modes. The simulation of the described approach is carried out on real data, the time results of the neural-like structure work and the accuracy results (MAPE, RMSE, R) are given. A comparison of the operation of the method with existing ones, such as Support vector regression, Classic linear SGTM neural-like structure, Linear regression (using Stochastic Gradient Descent), Random Forest, Multilayer Perceptron, AdaBoost are made. The advantages of the developed approach, in particular with regard to the highest accuracy among existing ones were experimentally established. © 2018 IEEE.Öğe Intelligent Irrigation System-Automation Using IoT Technology: A Review(Institute of Electrical and Electronics Engineers Inc., 2022) Mustafa, M.S.; Kutucu, H.The advent of the Internet of Things (IoT) devices in agriculture has had far-reaching effects. This article offers a wide-ranging survey of innovative technologys usefulness in farming. The Internet of Things, cloud computing, machine learning, and artificial intelligence are all cutting-edge technologies that explain smart agriculture. Smart farmings potential uses in food and fiber production, animal husbandry, and postharvest handling are explored. The effects of climate change on agricultural production are also examined. This research adds to the canon by reinforcing the difficulties of using smart technology in agriculture and bringing to light problems previously recognized by the current cutting-edge agricultural framework. The authors state that more research needs to be done to improve global food production, food management, and sustainability measures. They also highlight some gaps in the current research on how IoT can be used in smart farming. © 2022 IEEE.Öğe A mathematical model for finding the rainbow connection number(IEEE Computer Society, 2013) Nuriyeva, F.; Ugurlu, O.; Kutucu, H.The rainbow connection problem belongs to the class of NP-Hard graph theoretical problems. The rainbow connection of a connected graph G, denoted by rc(G), is the smallest number of colors that are needed in order to make G rainbow edge-connected. In this study, we present a new mathematical model for the rainbow connection problem. © 2013 IEEE.Öğe A method for constructing a barker-like sequences based on ideal ring bundles(CEUR-WS, 2019) Riznyk, O.; Kynash, Y.; Balych, B.; Vynnychuk, R.; Kret, I.; Kutucu, H.To date, all major reserves for improving the quality of wireless communication are almost exhausted. It is becoming clear to all wireless network designers and manufacturers that instead of using the standard principles of network efficiency enhancement, they should focus their efforts on implementing other principles of radio exchange by creating new code sequences. The Barker-like sequence is taken as the main sequence, and then each element of the main sequence is replaced by a direct or inverse additional Barker-like sequence, depending on whether there is zero or one in the main Barker-like sequence. An algorithm for the synthesis of Barker-like sequences using ideal ring bundles of different types is proposed. The method of constructing Barker combined signals and their autocorrelation functions are considered. The simulation of the obtained Barker-like sequences in the LabVIEW software environment is performed. Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0)Öğe ML-based Approach for Credit Risk Assessment Using Parallel Calculations(CEUR-WS, 2022) Hentosh, L.; Tsikalo, Y.; Kustra, N.; Kutucu, H.In banks and other credit organizations, the task of credit scoring often arises when making decisions on granting loans. The last one consists of making a reasoned decision based on information about the applicant, whether she should be granted a loan, and, if so, on what terms. This paper proposes the application of parallel calculations of the Random forest algorithm when solving the credit scoring task. This approach made it possible to reduce the time of model training and dataset processing significantly. Expectedly, when applying less data, the resulting acceleration and efficiency worsen. Using only 2500 entries, the execution time of the sequential algorithm is less than the parallel algorithm. The developed software was tested on three different processors: 4-core, 8-core, and 12-core, to evaluate the parallelization quality of data pre-processing. The classification algorithm is computationally complex and time-consuming, so we obtained practically the same acceleration for processing 5000 and 10000 records. With this amount of data, the 12-core processor gave the biggest gain in time when working with 12 threads. As a result, it is possible to have an acceleration of more than 6. This efficiency indicator of the proposed parallel algorithm can be significantly improved by varying the number of threads and considering the current trends in developing the multi-core architecture of computing systems. Also, using data without pre-processing, the following evaluation metrics were obtained: AUC=0.9 and Precision=0.845, and using data after pre-processing, these metrics were: AUC=0.86, Precision=0.89. © 2022 Copyright for this paper by its authors.Öğe Modeling of solar energy potential in Libya using an artificial neural network model(Institute of Electrical and Electronics Engineers Inc., 2016) Kutucu, H.; Almryad, A.In this work, we develop an artificial neural network model to predict the potential of solar power in Libya. We use multilayered, feed-forward, back-propagation neural networks for the mean monthly solar radiation using the data of 25 cities spread over Libya for the period of 6 years (2010-2015). Meteorological and geographical data (longitude, latitude, and altitude, month, mean sunshine duration, mean temperature, and relative humidity) are used as input to the network. The solar radiation is in the output layer of the network. The results show that the correlation coefficients between the ANN predictions and actual mean monthly global solar radiation for training and testing datasets are higher than 98%. Hence, the predictions from ANN model in locations where solar radiation data are not available has a high reliability. © 2016 IEEE.Öğe Network design problem with cut constraints(Springer International Publishing, 2018) Sharifov, F.; Kutucu, H.In this paper, we introduce a minimum cost network design problem with a given lower bound requirement for capacity of any cut. First, we consider the subproblem including the lower bound requirement only for each fundamental cut determined by deleting edges from some spanning tree. Then, it is shown that the simplex algorithm finds an optimal solution to standard LP-relaxation of the subproblem in the linear time of the number of edges. The simplex and minimum cut algorithms are used in a Branch-and-Cut type algorithm to pick up a solution to the minimum cost network design problem. © Springer International Publishing AG, part of Springer Nature 2018.Öğe Non-iterative neural-like predictor for solar energy in Libya(CEUR-WS, 2018) Tkachenko, R.; Kutucu, H.; Izonin, I.; Doroshenko, A.; Tsymbal, Y.In this paper, a new method for predicting the solar radiation potential in Libya was developed. It is constructed on the basis of the combined use of RBF and non-iterative paradigm of the artificial neural networks construction - the Successive Geometric Transformations Model. This method has the advantages of both approaches - the high prediction accuracy from RBF characteristics and fast non-iterative learning provided by the Successive Geometric Transformations Model. A series of practical experiments were conducted. The training model contained 1440 vectors of the monthly solar radiation, which recorded in 25 Libya's cities from 2010 to 2015. The test model contained 360 data's vectors. Comparison of the proposed method with existing ones is presented. The proposed method showed the best prediction results (MAPE, RMSE) compared to SVM, Linear Regression, the linear Neural-like structure of the Successive Geometric Transformation Model (SGTM), and the RBF based on the NLS SGTM. The proposed approach can be used in different areas, such as e-commerce, material science, images processing and others, especially in Big Data cases. © 2018 CEUR-WS. All rights reserved.Öğe ON THE SOLUTION APPROACHES OF THE BAND COLLOCATION PROBLEM(Turkic World Mathematical Soc, 2019) Kutucu, H.; Gursoy, A.; Kurt, M.; Nuriyev, U.This paper introduces the first genetic algorithm approach for solving the Band Collocation Problem (BCP) which is a combinatorial optimization problem that aims to reduce the hardware costs on fiber optic networks. This problem consists of finding an optimal permutation of rows of a given binary rectangular matrix representing a communication network so that the total cost of covering all 1's by Bands is minimum. We present computational results which indicate that we can obtain almost optimal solutions of moderately large size instances (up to 96 rows and 28 columns) of the BCP within a few seconds.Öğe ON THE SOLUTION APPROACHES OF THE BAND COLLOCATION PROBLEM(Isik University, 2019) Kutucu, H.; Gursoy, A.; Kurt, M.; Nuriyev, U.This paper introduces the first genetic algorithm approach for solving the Band Collocation Problem (BCP) which is a combinatorial optimization problem that aims to reduce the hardware costs on fiber optic networks. This problem consists of finding an optimal permutation of rows of a given binary rectangular matrix representing a communication network so that the total cost of covering all 1’s by Bands is minimum. We present computational results which indicate that we can obtain almost optimal solutions of moderately large size instances (up to 96 rows and 28 columns) of the BCP within a few seconds. © 2019. All rights reserved.Öğe The optimization of the bandpass lengths in the multi-bandpass problem(Springer Verlag, 2014) Kurt, M.; Kutucu, H.; Gürsoy, A.; Nuriyev, U.The Bandpass problem has applications to provide a cost reduction in design and operating telecommunication network. Given a binary matrix A m×n and a positive integer B called the Bandpass length, a set of B consecutive non-zero elements in any column is called a Bandpass. No two bandpasses in the same column can have common rows. The general Bandpass Problem consists of finding an optimal permutation of rows of the matrix A that produces the maximum total number of bandpasses having the same given bandpass length B in all columns. The Multi- Bandpass problem includes different bandpass lengths Bj in each column j of the matrix A, where j = 1,2, ?,n. In this paper, we propose an extended formulation for the Multi-Bandpass problem. A given Bj may not be always efficient bandpass lengths for the communication network. Therefore, it is important to find an optimal values of the bandpass lengths in the Multi-Bandpass problem. In this approach, the lengths in each destination are defined as zj and we present a model to find the optimal values of zj. Then, we calculate the approximate solution of this model using genetic algorithm for the problem instances which are presented in an online library. © Springer-Verlag Berlin Heidelberg 2014.Öğe THE RAINBOW CONNECTION PROBLEM: MATHEMATICAL FORMULATIONS(Charles Babbage Res Ctr, 2016) Kutucu, H.; Nuriyeva, F.; Ugurlu, O.The concept of rainbow connection was introduced by Chartrand et al. in 2008. The rainbow connection number, rc(G), of a connected graph G = (V, E) is the minimum number of colors needed to color the edges of E, so that each pair of the vertices in V is connected by at least one path in which no two edges are assigned the same color. The rainbow vertex-connection number, rvc(G), is the vertex version of this problem. In this paper, we introduce mixed integer programming models for both versions of the problem. We show the validity of the proposed models and test their efficiency using a nonlinear programming solver.
