Browsing by Author "Varol Altay E."
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Hybrid artificial neural network based on a metaheuristic optimization algorithm for the prediction of reservoir temperature using hydrogeochemical data of different geothermal areas in Anatolia (Turkey)(Elsevier Ltd, 2022) Varol Altay E.; Gurgenc E.; Altay O.; Dikici A.Due to the increase in the changes in global climate in recent years and the depletion of fossil fuels, the interest in renewable energy sources in many developed countries is increasing day by day. Among the renewable energy sources, geothermal energy has an important place because it can be used both in electricity production and directly as heat energy. Before using geothermal fluids, it is necessary to determine their properties by making detailed geological studies and thus to determine the most suitable drilling location. These processes are very costly, time-consuming, and require special equipment. Such disadvantages can be eliminated by using machine learning methods. In this study, the machine learning methods developed for the classification approach were used to predict the purpose of the geothermal waters with the help of the geothermal data set obtained from different regions. In this study, naïve Bayes classifier, K-nearest neighbor, linear discrimination analysis, binary decision tree, support vector machine, and artificial neural network, which are widely used in the literature, were used. In addition, promising results were obtained by designing a hybrid metaheuristic artificial neural network model. While an accuracy in traditional machine learning methods between 71% and 82% was obtained, a 91.84% accuracy was obtained in the model proposed. © 2022 Elsevier LtdItem Hybrid Archimedes optimization algorithm enhanced with mutualism scheme for global optimization problems(Springer Nature, 2023) Varol Altay E.Archimedes optimization algorithm (AOA) is a recent metaheuristic method inspired by the Archimedes principle, which is the law of physics. Like other metaheuristic methods, it suffers from the disadvantages of being stuck in local areas, suffering from weak exploitation abilities, and an inability to maintain a balance between exploration and exploitation. To overcome these weaknesses, a new hybrid Mutualism Archimedes Optimization Algorithm (MAOA) method has been proposed by combining the AOA and the mutation phase in the Symbiosis organism search (SOS) method. SOS algorithm is known for its exploitation ability. With the mutation phase, it has been used to improve local search for swarm agents, help prevent premature convergence and increase population diversity. To verify the applicability and performance of the proposed algorithm, extensive analysis of standard benchmark functions, CEC’17 test suites, and engineering design problems were performed. The proposed method is compared with the recently emerged and popular AOA, SOS, Harris Hawks Optimization (HHO), COOT Optimization Algorithm (COOT), Aquila Optimizer (AO), Salp Swarm Algorithm (SSA), and Multi-Verse Optimization (MVO) methods, and statistical analyses were performed. The results obtained from the experiments show that the proposed MAOA method has superior global search performance and faster convergence speed compared to AOA, SOS, and other recently emerged and popular metaheuristic methods. Furthermore, this study compares MAOA to five well-established and recent algorithms constructed using various metaheuristic methodologies utilizing nine benchmark datasets to assess the general competence of MAOA in feature selection. Therefore, the proposed method is considered to be a promising optimization method for real-world engineering design problems, global optimization problems, and feature selection. © 2022, The Author(s), under exclusive licence to Springer Nature B.V.Item Assessment of Grey Wolf Optimizer and Its Variants on Benchmark Functions(Springer Science and Business Media Deutschland GmbH, 2023) Varol Altay E.; Altay O.One of the most current metaheuristic swarm intelligence algorithms is the Grey Wolf Optimizer (GWO). Since the number of parameters is small and there is no need for information during the first search, GWO has been adapted to different optimization problems, giving it superiority over other metaheuristic methods. At the same time, it is easy to use, simple, scalable, and adaptable, with its unique ability to provide the ideal balance throughout the search, leading to positive convergence. As a result, the GWO has lately attracted a large research audience from a variety of areas in a very short amount of time. However, it has some disadvantages, such as a slow convergence rate, low sensitivity, and so on, which are seen in the vast majority of metaheuristic methods. Various versions of the current GWO have been proposed to eliminate them. In this article, GWO, improved GWO (IGWO), and augmented GWO (AGWO) methods are examined, and the performances of these methods are discussed in CEC’20 functions and analyzed statistically. The results of the studies demonstrated that IGWO outperformed standard GWO and AGWO. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item A novel hybrid multilayer perceptron neural network with improved grey wolf optimizer(Springer Science and Business Media Deutschland GmbH, 2023) Altay O.; Varol Altay E.The multilayer perceptron (MLP), a type of feed-forward neural network, is widely used in various artificial intelligence problems in the literature. Backpropagation is the most common learning method used in MLPs. The gradient-based backpropagation method, which is one of the classical methods, has some disadvantages such as entrapment in local minima, convergence speed, and initialization sensitivity. To eliminate or minimize these disadvantages, there are many studies in the literature that use metaheuristic optimization methods instead of classical methods. These methods are constantly being developed. One of these is an improved grey wolf optimizer (IMP-GWO) proposed to eliminate the disadvantages of the grey wolf optimizer (GWO), which suffers from a lack of search agent diversity, premature convergence, and imbalance between exploitation and exploration. In this study, a new hybrid method, IMP-GWO-MLP, machine learning method was designed for the first time by combining IMP-GWO and MLP. IMP-GWO was used to determine the weight and bias values, which are the most challenging parts of the MLP training phase. The proposed IMP-GWO-MLP was applied to 20 datasets consisting of three different approximations, eight regression problems, and nine classification problems. The results obtained have been suggested in the literature and compared with the gradient descent-based MLP, commonly used GWO, particle swarm optimization, whale optimization algorithm, ant lion algorithm, and genetic algorithm-based MLP methods. The experimental results show that the proposed method is superior to other state-of-the-art methods in the literature. In addition, it is thought that the proposed method can be modeled with high success in real-world problems. © 2022, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.Item Average Localization Error Prediction for 5G Networks: An Investigation of Different Machine Learning Algorithms(Springer, 2024) Altay O.; Erel-Özçevik M.; Varol Altay E.; Özçevik Y.In the realm of today’s networking technologies, user localization has been a formidable challenge for recent applications. There are different approaches in pursuit of heightened position detection of an end-user with the help of GPS, Wi-Fi fingerprint and 5G equipment. However, these approaches require both deployment and maintenance costs because of equipment establishment for position tracking. Moreover, they are not capable of minimizing the localization error, especially for indoor scenarios to track the indoor position of an end-user. Hence, there is an urgent need to delve deeper into innovative approaches to drive further advancements in user localization. In response, Machine Learning (ML) approaches have recently been widely adapted to predict the localization of end-users with minimum error. More specifically, average localization error (ALE) of an end-user can be predicted in a cost-effective way by using proper data and ML methods. For this purpose, we have investigated different ML approaches to get an accurate ALE prediction scheme for 5G networks with mobile end-users. Accordingly, an existing dataset is utilized to generate localization data of end-users in which the ALE is directly calculated by Received Signal Strength Indicator. Moreover, three different normalization approaches are applied for the overarching goal of increased data quality. Consequently, six different ML algorithms, including Linear regression, support vector machine with three different kernels, Gaussian process, and ensemble least-squares boosting (LSBoost) are evaluated with respect to a set of evaluation criteria including R, R2, RMSE, and MAE. The evaluation outcomes emphasize that ensemble LSBoost method, in the context of localization prediction, outperforms the other approaches and is sufficient to yield a viable learning strategy for ALE prediction. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.