Yazar "Arslan, Hilal" seçeneğine göre listele
Listeleniyor 1 - 7 / 7
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Betweenness centrality in sparse real world and wireless multi-hop networks(Springer Science and Business Media Deutschland GmbH, 2022) Tuzcu, Atakan; Arslan, HilalGraphs are one of the compact ways to represent information about real-life and intelligent system networks like wireless sensor networks. Betweenness centrality is an important network measure that evaluates the significance of a node based on the shortest paths and is widely used in biological, social, transportation, complex, and communication networks. In this study, we implement an efficient algorithm computing betweenness centrality of nodes for real-world and wireless multi-hop networks. Large sparse graphs are stored using compressed sparse row storage format and modified version of Dijkstra’s algorithm is used to compute shortest paths. We conduct a comprehensive experimental study on real-world networks as well as wireless sensor networks that are state-of-the-art technologies for different applications such as intelligence structures, industrial and home automation as well as health care. We evaluate the effect of network dimension on the time needed to compute betweenness centrality. Experimental results demonstrate that computation time required to compute betweenness centrality varies from 0.9 to 52.5 s when the number of vertices changes from 10,000 to 60,000. We also observe that the proposed algorithm efficiently computes betweenness centrality for networks coming from machine learning, power network, and networks obtained from optimization problems as well as computational fluid dynamics. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Öğe Betweenness Centrality in Sparse Real World and Wireless Multi-hop Networks(Springer Science and Business Media Deutschland GmbH, 2022) Tuzcu, Atakan; Arslan, HilalGraphs are one of the compact ways to represent information about real-life and intelligent system networks like wireless sensor networks. Betweenness centrality is an important network measure that evaluates the significance of a node based on the shortest paths and is widely used in biological, social, transportation, complex, and communication networks. In this study, we implement an efficient algorithm computing betweenness centrality of nodes for real-world and wireless multi-hop networks. Large sparse graphs are stored using compressed sparse row storage format and modified version of Dijkstra’s algorithm is used to compute shortest paths. We conduct a comprehensive experimental study on real-world networks as well as wireless sensor networks that are state-of-the-art technologies for different applications such as intelligence structures, industrial and home automation as well as health care. We evaluate the effect of network dimension on the time needed to compute betweenness centrality. Experimental results demonstrate that computation time required to compute betweenness centrality varies from 0.9 to 52.5 s when the number of vertices changes from 10,000 to 60,000. We also observe that the proposed algorithm efficiently computes betweenness centrality for networks coming from machine learning, power network, and networks obtained from optimization problems as well as computational fluid dynamics. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Öğe A comparative study on COVID-19 prediction using deep learning and machine learning algorithms: a case study on performance analysis(2022) Arslan, Hilal; Er, OrhanCOVID-19 disease has been the most important disease recently and has affected serious number of people in the world. There is not proven treatment method yet and early diagnosis of COVID-19 is crucial to prevent spread of the disease. Laboratory data can be easily accessed in about 15 minutes, and cheaper than the cost of other COVID-19 detection methods such as CT imaging and RT-PCR test. In this study, we perform a comparative study for COVID-19 prediction using machine learning and deep learning algorithms from laboratory findings. For this purpose, nine different machine learning algorithms including different structures as well as deep neural network classifier are evaluated and compared. Experimental results conduct that cosine k-nearest neighbor classifier achieves better accuracy with 89% among other machine learning algorithms. Furthermore, deep neural network classifier achieves an accuracy of 90.3% when one hidden layer including 60 neurons is used to detect COVID-19 disease from laboratory findings data.Öğe A new COVID-19 detection method from human genome sequences using CpG island features and KNN classifier(Elsevier - Division Reed Elsevier India Pvt Ltd, 2021) Arslan, Hilal; Arslan, HasanVarious viral epidemics have been detected such as the severe acute respiratory syndrome coronavirus and the Middle East respiratory syndrome coronavirus in the last two decades. The coronavirus disease 2019 (COVID-19) is a pandemic caused by a novel betacoronavirus called severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). After the rapid spread of COVID-19, many researchers have investigated diagnosis and treatment for this terrifying disease quickly. Identifying COVID-19 from the other types of coronaviruses is a difficult problem due to their genetic similarity. In this study, we propose a new efficient COVID-19 detection method based on the K-nearest neighbors (KNN) classifier using the complete genome sequences of human coronaviruses in the dataset recorded in 2019 Novel Coronavirus Resource. We also describe two features based on CpG island that efficiently detect COVID-19 cases. Thus, genome sequences including approximately 30,000 nucleotides can be represented by only two real numbers. The KNN method is a simple and effective non-parametric technique for solving classification problems. However, performance of the KNN depends on the distance measure used. We perform 19 distance metrics investigated in five categories to improve the performance of the KNN algorithm. Some efficient performance parameters are computed to evaluate the proposed method. The proposed method achieves 98.4% precision, 99.2% recall, 98.8% F-measure, and 98.4% accuracy in a few seconds when any L1 type metric is used as a distance measure in the KNN. (c) 2020 Karabuk University. Publishing services by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Öğe A parallel fully dynamic iterative Bio-Inspired Shortest Path Algorithm(Springer Heidelberg, 2020) Arslan, HilalWe propose a fully dynamic bio-inspired parallel algorithm for solving the shortest path problem on dynamically changing graphs based on Physarum Solver, which is an amoeba shortest path algorithm. Although many sequential algorithms exist for solving dynamic shortest path problem, there are only few parallel algorithms, most of which identify the set of vertices affected by the dynamic changes. However, when the graph size becomes large, the process of determining affected vertices is time-consuming. In fact, when the percentage of changing edges is large, most of the studies in the literature are infeasible. The proposed algorithm is able to identify affected vertices and reconstruct them spontaneously in parallel. Moreover, it is designed to be suitable for dynamically changing graphs since it uses the information arising in earlier iterations. Thus, it computes effectively dynamic shortest path even if percentage of changing edges is large. At each iteration of the proposed algorithm, a sparse linear system of equations needs to be solved, which is the most time-consuming and challenging step of the algorithm especially when the problem size is large. We propose a parallel preconditioned iterative method for solving those sparse linear systems. The proposed preconditioner is specifically designed based on the properties of the coefficient matrix of those linear systems, and the effectiveness of the proposed preconditioner is compared against other state-of-the-art preconditioners on dynamic graphs. The proposed algorithm is evaluated using three different large graph models representing diverse real-life applications on a parallel multicore cluster. The parallel scalability of the proposed algorithm is compared against Delta-stepping, which is the most representative parallel implementation of Dijkstra's algorithm in the Parallel Boost Graph Library.Öğe Parallel identification of central nodes in wireless multi-hop networks(IEEE, 2020) Eliiyi, Deniz Türsel; Arslan, Hilal; Akram, Vahid Khalilpour; Uğurlu, OnurA wireless multi-hop network is a collection of nodes that communicate by message passing over multiple links. Sending a message to a remote node can consume some energy from all intermediary nodes. In a network, the nodes with minimum distance to all other nodes are called Jordan central nodes. Selecting the central nodes as sink or base station can considerably reduce the overall energy consumption and increase the network life time. This paper proposes a new parallel algorithm to find all central nodes of a network by finding BFS trees of a subset of nodes. The roots of the trees with smallest height are selected as Jordan central nodes. After finding each tree the algorithm eliminates some nodes from the search space. Available processors construct the BFS tree for different nodes in parallel and eliminate a group of unvisited nodes after creating each tree. The implementation results of the algorithm using different number of processors on topologies with up to 250 nodes showed that the proposed algorithm can find all central nodes by examining less than 20% of nodes in less than 0.034 seconds.Öğe A progressive search algorithm for the minimum hitting set problem(2021) Arslan, Hilal; Uğurlu, Onur; Akram, Vahid Khalilpour; Eliiyi, Deniz TürselGiven a finite universe and a collection of the subsets of the universe, the minimum hitting set of thecollection is the smallest subset of the universe that has non-empty intersection with each set in thecollection. Finding the minimum hitting set is an NP-Hard problem that has many real worldapplications. In this study, we propose a progressive search-based approach to find the minimumhitting set of a given collection. The algorithm starts searching for the hitting sets of size 1 andincrease the expected size of the minimum hitting set by a factor of d. After each unsuccessful search,it increases the expected size by d and generate the candidate sets with the expected size. After eachsuccessful search, the algorithm takes the average of last unsuccessful and successful searches andcontinue the searching with the new expected size. The algorithm terminates when the detectedupper bound coincides with the detected lower bound. The effect of different values for d on theperformance of the algorithm has been experimented on various data sets. Experimental resultsreveal that the proposed method effectively computes the minimum hitting set on real-world dataand random dataset.
