Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/395
Browse
9 results
Search Results
Article Citation - WoS: 1A Comprehensive Analysis of Acoustic Emission Signals To Distinguish the Different Damage Types for Fiber-Reinforced Polymers: A Review(Wiley, 2025-12-03) Yilmaz, CagatayFiber-reinforced polymers (FRP) attract the attention of key industries, such as aerospace, wind energy, and automotive, as they can reduce the weight of structural components without compromising their mechanical properties. Due to FRP's anisotropic and non-homogeneous structure, their failure under different loading conditions and the corresponding failure mechanisms must be investigated. One method that progressively monitors the failure of FRP underload is Acoustic Emission (AE). AE can register the elastic stress waves in the form of digitized waveforms, released by the discontinuous events that occur in the FRP under load. These discontinuities can be clustered and identified as transverse cracking, fiber/matrix interface debonding, delamination, and fiber failure by analyzing the AE waveforms. Recently, numerous clustering approaches using machine learning algorithms, along with the varying features of AE waveforms, have been developed and are being used. These algorithms include supervised and unsupervised clustering, deep learning algorithms, and neural network methods, among others. While supervised algorithms require a training dataset to classify AE signals, unsupervised algorithms can perform clustering without training datasets. Deep learning and neural network algorithms can train themselves to cluster data, but they may require a significant amount of computer power when the dataset is large. This review paper provides comprehensive information on the clustering algorithm, along with the AE wave features, the range of features for different damage types, and the type of reinforcer.Conference Object The Effect of Different Classifiers on Recursive Cluster Elimination in the Analysis of Transcriptomic Data(Institute of Electrical and Electronics Engineers Inc., 2023-10-11) Bulut, Nurten; Bakir-Güngör, Burcu; Qaqish, Bahjat F.; Yousef, MalikGene expression data with limited sample size and a large number of genes are frequently encountered in genetic studies. In such high-dimensional data, identification of genes that distinguish between disease states is a challenging task. Feature selection (FS) is a useful approach in dealing with high dimensionality. Support Vector Machines Recursive Cluster Elimination (SVM-RCE) is a technique for FS in high-dimensional data. The SVM-RCE approach has been utilized for identification of clusters of genes whose expression levels correlate with pathological state. A key step in SVM-RCE is the use of an SVM classifier to assign an area under the curve (AUC) score to each gene cluster based on its ability to predict class labels. In this study, we investigate the use of alternative classifiers in the cluster-scoring step. Specifically, we compare Support Vector Machines, Random Forest, XgBoost, Naive Bayes, and linear logistic regression. In addition to AUC score performance evaluation, the algorithms are compared in terms of the number of selected genes at different levels of clustering and in terms of the running time. © 2023 Elsevier B.V., All rights reserved.Article Citation - WoS: 52Citation - Scopus: 69Spectrum-Aware Bio-Inspired Routing in Cognitive Radio Sensor Networks for Smart Grid Applications(Elsevier Science Bv, 2017-03) Fadel, E.; Faheem, M.; Gungor, V. C.; Nassef, L.; Akkari, N.; Malik, M. G. A.; Akyildiz, I. F.Cognitive radio sensor networks (CRSNs) have been proposed to serve as a reliable, robust, and efficient communications infrastructure that can address both the existing and future energy management requirements of the smart grid. The existing and envisioned applications of CRSN-based smart grid include substation automation, overhead transmission line monitoring, home energy management, advanced metering infrastructure, wide-area situational awareness, demand response, outage management, distribution automation, asset management. To realize these applications, in this paper, honey bee mating optimization-based routing and cooperative channel assignment algorithms have been proposed. The developed framework significantly decreases the probability of packet loss and preserves high link quality among sensor nodes in harsh smart grid spectrum environments. The proposed approach performance has been evaluated in terms of packet delivery ratio, delay, and energy consumption demonstrating that it has successfully addressed the QoS requirements of most of the SG applications presented. (C) 2017 Elsevier B.V. All rights reserved.Article Citation - WoS: 74Citation - Scopus: 89QERP: Quality-Of (QoS) Aware Evolutionary Routing Protocol for Underwater Wireless Sensor Networks(IEEE-Inst Electrical Electronics Engineers Inc, 2018-09) Faheem, Muhammad; Tuna, Gurkan; Gungor, Vehbi CagriQuality-of-service (QoS) aware reliable data delivery is a challenging issue in underwater wireless sensor networks (UWSNs). This is clue to impairments of the acoustic transmission caused by excessive noise, extremely long propagation delays, high bit error rate, low bandwidth capacity, multipath effects, and interference. To address these challenges, meet the commonly used UWSN performance indicators, and overcome the inefficiencies of the existing clustering-based routing schemes, a novel QoS aware evolutionary cluster based routing protocol (QERP) has been proposed for UWSN-based applications. The proposed protocol improves packet delivery ratio, and reduces average end-to-end delay and overall network energy consumption. Our comparative performance evaluations demonstrate that QERP is successful in achieving low network delay, high packet delivery ratio, and low energy consumption.Article Citation - WoS: 5Citation - Scopus: 4Investigating Strain Rate Effects on Damage Mechanisms in Hybrid Laminated Composites Using Acoustic Emission(Elsevier Sci Ltd, 2025-12) Gulsen, Abdulkadir; Kolukisa, Burak; Etcil, Mustafa; Caliskan, Umut; Zafar, Hafiz Muhammad Numan; Demirbas, Munise Didem; Bakir-Gungor, BurcuHybrid composites, which combine distinct fiber types such as carbon, basalt, and aramid, provide a synergistic balance of strength, stiffness, impact resistance, and energy dissipation, making them appealing for critical applications in aerospace, automotive, and other high-performance industries. Monitoring damage progression in these composites is vital for ensuring structural integrity and preventing catastrophic failures. Acoustic emission (AE) serves as a powerful, noninvasive technique for real-time structural health monitoring, capturing the transient stress waves generated when damage events occur. This study utilizes AE to examine the influence of strain rate on damage modes in carbon/basalt/aramid hybrid composites under three-point bending. An unsupervised feature selection based on Laplacian scores is employed to identify the most relevant AE features with damage modes, while SHapley Additive Explanations (SHAP) are used to evaluate the correlation between AE features and strain rates. The correlation analysis results indicate that peak frequency (PF) serves as a key indicator, demonstrating significant shifts at higher strain rates. Gaussian Mixture Model (GMM) clustering is used to analyze hybrid composites by examining clustered AE signals based on selected features identified through Laplacian scores, with Silhouette scores employed to determine the optimal number of clusters. This study highlights the role of AE in understanding fiber interactions and damage evolution, offering valuable insights into the mechanical performance and optimization of carbon/basalt/aramid hybrid composite structures.Conference Object Enhancing Gene Expression Data Analysis Through SVM-Based Recursive Cluster Elimination and Weighted Center Approaches(Avestia Publishing, 2024-08) Yousef, Malik; Bulut, Nurten; Gungor, Burcu Bakir; Qaqish, Bahjat F.The complexity and high dimensionality of gene expression data pose significant challenges for effective feature selection and accurate classification in bioinformatics. This study introduces two novel algorithms, Support Vector Machine-Recursive Cluster Elimination (SVM-RCE) and its advanced version, SVM-RCE with Center Weights (SVM-RCE-CW), designed to optimize feature selection by leveraging clustering techniques and machine learning models. Both algorithms aim to reduce the feature space, thereby enhancing the interpretability and performance of classification models. We present a comprehensive comparison of these methods against traditional feature selection techniques, demonstrating their efficacy in achieving significant dimensionality reduction while maintaining or improving classification accuracy in several gene expression datasets. © 2024 Elsevier B.V., All rights reserved.Conference Object Citation - Scopus: 2Effect of Recursive Cluster Elimination With Different Clustering Algorithms Applied to Gene Expression Data(Institute of Electrical and Electronics Engineers Inc., 2023-10-11) Kuzudisli, Cihan; Bakir-Güngör, Burcu; Qaqish, Bahjat F.; Yousef, MalikFeature selection (FS) is an effective tool in dealing with high dimensionality and reducing computational cost. Support Vector Machines-Recursive Cluster Elimination (SVM-RCE) is one of several algorithms that have been developed for FS in high dimensional data. SVM-RCE involves a clustering step which originally is k-means. Using various performance metrics, three alternative algorithms are evaluated in this context; k-medoids, Hierarchical Clustering (HC), and Gaussian Mixture Model (GMM). Comparisons will be carried out on five publicly available gene expression datasets. The results show that k-means in SVM-RCE obtains higher performance than other tested algorithms in terms of classification performance. Additionally, HC shows a similar performance to k-means. Our findings show superiority of using k-means. This study can contribute to the development of SVM-RCE with different variations, leading to decrease in the number of selected genes, and an increase in prediction performance. © 2023 Elsevier B.V., All rights reserved.Article Citation - WoS: 64Citation - Scopus: 76EDHRP: Energy Efficient Event Driven Hybrid Routing Protocol for Densely Deployed Wireless Sensor Networks(Academic Press Ltd- Elsevier Science Ltd, 2015-12) Faheem, Muhammad; Abbas, Muhammad Zahid; Tuna, Gurkan; Gungor, Vehbi CagriEfficient management of energy resources is a challenging research area in Wireless Sensor Networks (WSNs). Recent studies have revealed that clustering is an efficient topology control approach for organizing a network into a connected hierarchy which balances the traffic load of the sensor nodes and improves the overall scalability and the lifetime of WSNs. Inspired by the advantages of clustering techniques, we have three main contributions in this paper. First, we propose an energy efficient cluster formation algorithm called Active Node Cluster Formation (ANCF). The core aim to propose ANCF algorithm is to distribute heavy data traffic and high energy consumption load evenly in the network by offering unequal size of clusters in the network. The developed scheme appoints each cluster head (CH) near to the sink and sensing event while the remaining set of the cluster heads (CHs) are appointed in the middle of each cluster to achieve the highest level of energy efficiency in dense deployment. Second, we propose a lightweight sensing mechanism called Active Node Sensing Algorithm (ANSA). The key aim to propose the ANSA algorithm is to avoid high sensing overlapping data redundancy by appointing a set of active nodes in each cluster with satisfy coverage near to the event. Third, we propose an Active Node Routing Algorithm (ANRA) to address complex inter and intra cluster routing issues in highly dense deployment based on the node dominating values. Extensive experimental studies conducted through network simulator NCTUNs 6.0 reveal that our proposed scheme outperforms existing routing techniques in terms of energy efficiency, end-to-end delay and data redundancy, congestion management and setup robustness. (C) 2015 Elsevier Ltd. All rights reserved.Article Citation - WoS: 12Citation - Scopus: 19An Approximate Spectral Clustering Ensemble for High Spatial Resolution Remote-Sensing Images(IEEE-Inst Electrical Electronics Engineers Inc, 2015-05) Tasdemir, Kadim; Moazzen, Yaser; Yildirim, IsaUnsupervised clustering of high spatial resolution remote-sensing images plays a significant role in detailed land-cover identification, especially for agricultural and environmental monitoring. A recently promising method is approximate spectral clustering (SC) which enables spectral partitioning for large datasets to extract clusters with distinct characteristics without a parametric model. It also facilitates the use of various information types via advanced similarity criteria. However, it requires an empirical selection of a similarity criterion optimal for the corresponding application. To address this challenge, we propose an approximate SC ensemble (ASCE2) which fuses partitionings obtained by different similarity representations. Contrary to existing spectral ensembles for remote-sensing applications, the proposed ASCE2 employs neural gas quantization instead of random sampling, advanced similarity criteria instead of traditional distance-based Gaussian kernel with different decay parameters, and a two-level ensemble. We evaluate the proposed ASCE2 with three measures (accuracy, adjusted Rand index, and normalized mutual information) using five remote-sensing images, two of which are commonly available. We apply the ASCE2 in two applications for agricultural monitoring: 1) land-cover identification to determine orchard fields using a WorldView-2 image (0.5-m spatial resolution) and 2) finding lands in good agricultural condition using multitemporal RapidEye images (5-m spatial resolution). Experimental results indicate a significant betterment of the resulting partitionings obtained by the proposed ensemble, with respect to the evaluation measures in these applications.