Bilgisayar Mühendisliği Bölümü Yayın Koleksiyonu

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/11494/296

Listeleniyor 1 - 20 / 53

BCI for mobile devices: Time-frequency and representation learning analysis of mobile gesture tasks
(Institute of Electrical and Electronics Engineers Inc., 2025) Yılmaz, Çagatay Murat; Ulu, Ahmet; Demirbaş, Gülşah
Mobile devices are central to daily life, yet their interaction remains primarily limited to touch and voice, which may be impractical in restricted mobility scenarios. Brain-computer interfaces (BCIs) offer a promising alternative by enabling device control through neural activity. Although motor imagery (MI) BCIs have been extensively studied, MI signals for mobile-specific gestures remain relatively underexplored. This study addresses this gap by investigating the effectiveness of combining time-frequency transformations with deep learning for classifying MI of mobile gestures. Using the MI-BMPI dataset-comprising EEG recordings of participants imagining tapping and swiping, and among the few publicly available datasets of its kind-we applied Short Time Fourier Transform, Continuous Wavelet Transform, Stockwell Transform, and Hilbert-Huang Transform as inputs to architectures including ResNeSt-50d, HRNet-W18, ConvNeXt-Base, DeiT-Base, Swin Transformer-Tiny, and RegNetX-002. In the classification of tapping and swiping MI tasks, performance across different subjects ranged from 0.61 to 0.947 when averaged over five random seeds, while individual subject models achieved results between 0.708 and 0.975. These promising outcomes are consistent with other evaluation metrics. Overall, this work provides the first systematic evaluation of mobile gesture MI-BCIs, demonstrating that time-frequency representations coupled with deep learning enable robust EEG-based mobile interaction and laying the groundwork for the development of future neuroadaptive technologies.
Comparison of the estimation methods for the parameters of exponentiated reduced kies distribution
(Süleyman Demirel Üniversitesi, 2018) Akgül, Fatma Gül
In this paper, we consider the estimation for the parameters ofexponentiated reduced Kies (ERK) distribution using maximum likelihood (ML),least squares (LS), weighted least squares (WLS), Cramér-von Mises (CM), AndersonDarling (AD) and right-tail Anderson Darling (RAD) methods. The performances ofthe estimators are compared via Monte-Carlo simulation study for differentparameter settings and different sample sizes. Finally, a real data set is analyzed forthe implementation of the proposed methods.
Resilience, intolerance of uncertainty, future anxiety and mental well being among young researchers
(Springer, 2025) Satıcı, Seydi Ahmet; Kütük, Hasan; Okur, Sinan; Demirci, İbrahim; Deniz, M. Engin; Satıcı, Begüm; Kayiş, Ahmet Rıfat; Aksu, M. Çağrı; Körün, Ali Berke; Okur, Ezgi; Yılmaz, Fatma Betül; Bırni, Gaye; Kaya, Yağmur; Aktepe, Zahide Gül
Young researchers face significant psychological challenges that may jeopardize their mental health. In order to protect and support the psychological health of young researchers, the determinants of mental well-being should be examined. This study aims to examine the serial mediating roles of intolerance of uncertainty and future anxiety in the relationship between resilience and mental well-being in young researchers. The study sample consisted of 259 female and 146 male doctoral students (Mage = 30.385, SD = 3.062). The research findings revealed that resilience, intolerance of uncertainty, and future anxiety are important predictors of young researchers' mental well-being. The results indicate that resilience positively predicts mental well-being, while intolerance of uncertainty and future anxiety play a partially serial mediating role in this relationship. This finding means that young researchers with low resilience experience more intolerance of uncertainty, which increases their future anxiety and thus leads to a decrease in mental well-being. This result emphasizes the importance of developing resilience in young researchers to reduce the negative effects of uncertainty and anxiety on mental health. Being resilient to navigate uncertain situations and reduce future anxiety is a protective factor for the mental well-being of young researchers. All these findings expand the understanding of the protection of the mental health of young researchers and offer important implications.
Region-specific topographic representations for deep learning-based brain-computer interfaces
(Institute of Electrical and Electronics Engineers Inc., 2025) Yılmaz, Çağatay Murat; Ulu, Ahmet
Electroencephalography (EEG) signals are widely used across disciplines and are promising for future medical and technological applications. However, current EEG analysis methods often fall short in classification accuracy, limiting progress in brain-computer interfaces (BCIs), early diagnosis of neurological disorders, and AI-driven health technologies. This paper proposes a region-specific spatial-spectral topographic mapping approach to enhance EEG-based classification using deep learning. The core objective is to improve discriminative feature learning by focusing only on brain regions relevant to motor-related activity and utilizing spectral representations (alpha, beta, and alpha/beta ratio). Pre-trained deep learning architectures-ResNeSt-50d, HRNet-W18, and ConvNeXt-Base-were fine-tuned on topographic images generated from the BCI Competition IV dataset 2a for classifying left- and right-hand motor imagery tasks. The proposed method achieved classification accuracies exceeding 75% when data from multiple sessions were aggregated. These results demonstrate the effectiveness of spatially and spectrally informed topographic representations for robust EEG-based BCI systems and lay a foundation for integrating EEG signals into advanced artificial intelligence applications.
Classification of medical imaging technologies: Results from Türkiye
(BioMed Central Ltd, 2025) Temiz, Hakan; Kara, Tuncay
Background: Regional disparities in access to medical diagnostic imaging technologies (MDITs) present a significant barrier to achieving health equity, particularly in developing countries. Understanding how these technologies are distributed and utilized is essential for informing equitable health policy. Method: This study examines the distribution and utilization of MDITs across Türkiye’s 12 NUTS regions using a hierarchical clustering approach. Unlike previous studies, the analysis incorporates both technological capacity and utilization (CaU) variables, evaluated jointly and independently. Imaging modalities are also stratified based on their technological complexity and investment requirements to capture nuanced regional patterns. Results: Findings indicate that although Türkiye demonstrates an overall balanced distribution of MDITs, notable regional disparities in utilization efficiency remain. Regions exhibiting similar usage patterns tend to cluster together irrespective of geographic proximity. Interestingly, the clusters often transcend geographical proximity; regions located at opposite ends of the country tend to cluster on the basis of similar utilization patterns. This may suggest that disparities between administrative centers and rural areas are less pronounced than previously assumed. These patterns imply that institutional capacity, healthcare workforce distribution, and demographic demand may have a stronger influence on utilization than spatial location. Conclusion: The study highlights a disconnect between capacity and actual use of diagnostic imaging technologies, underscoring the need for targeted policy interventions. It also suggests that regional utilization patterns may align more with functional similarities than with geographic proximity. Moreover, analyzing technological capacity and utilization variables separately—rather than as a combined index—yielded more transparent and objective insights into regional disparities. These findings contribute to optimizing health resource allocation and support evidence-based policymaking aimed at advancing equitable access to diagnostic services, aligning with Türkiye’s commitment to universal health coverage.
Enhancing the resolution of historical Ottoman texts using deep learning-based super-resolution techniques
(International Information and Engineering Technology Association, 2023) Temiz, Hakan
The Ottoman Empire's extensive archives hold valuable insights into centuries of history, necessitating the preservation and transfer of this rich heritage to future generations. To facilitate access and analysis, numerous digitization efforts have been undertaken to transform these valuable resources into digital formats. The quality of digitized documents directly impacts the success of tasks such as text search, analysis, and character recognition. This study aims to enhance the resolution and overall image quality of Ottoman archive text images using four deep learning-based super-resolution (SR) algorithms: VDSR, SRCNN, DECUSR, and RED-Net. The performance of these algorithms was assessed using SSIM, PSNR, SCC, and VIF image quality measures (IQMs) and evaluated in terms of human visual system perception. All SR algorithms achieved promising IQM scores and a significant improvement in image quality. Experimental results demonstrate the potential of deep learning-based SR techniques in enhancing the resolution of historical Ottoman text images, paving the way for more accurate character recognition, text processing, and analysis of archival documents.
SI and binary prefixes: Clearing the confusion
(Association for Computing Machinery, 2023) Temiz, Hakan
THE INTERNATIONAL SYSTEM of Units (SI), organized by the International Bureau of Weights and Measures (BIPM), defines several prefixes to denote the quantities between 10–24 and 10+24, given in Table 1. It allows expression of very large or small quantities of all units of measurement in common scientific notation accepted by the international community. For example, the prefixes kilo (k), mega (M), giga (G), and tera (T) indicate 103n, for n= 1, 2, 3, and 4, respectively. However, when expressing data quantities and memory addresses, SI prefixes (SIPs) are misused to indicate binary multiples. In this non-standard adoption, for a number n, where n=1, 2, …, and 8, each prefix specifies 210n instead of their original values (103n). Almost everyone, including scientists and engineers, thinks that this non-standard acceptance is correct. On the other hand, the prefixes for binary multiples were already defined by the relevant standards approximately 23 years ago. This nonstandard notation confuses at the very least or can cause rather serious problems. Hence, all parties should immediately abandon this misconception and disseminate the correct information to everyone.
Corpus-Assisted multidimensional AI applications in language teaching and an intelligent virtual speaker
(IGI Global, 2023) Özbay, Ali Şükrü; Hoşoğlu, Ayşenur; Uzuner, Buse; Öztürk, Ercüment
The authors propose an evidence-based virtual AI application geared to using a successful implementation of AI for language learners and language teaching professionals and a virtual speaking tool that will serve as a foundation for further research and experimentation. The study promotes interdisciplinary collaboration among corpus linguists, language educators, and computer engineers for fostering the skills and the competences and the exchange of ideas and a shared vision for the future of AI-assisted language learning. The main objective of this action plan is to propose an application in which language teaching professionals and language learners will engage in English-speaking practice through a multidimensional (MD) environment incorporating cubic images and bilateral sound perception (hearing and speaking). The AI application will assign tasks to multidimensional characters using realistic sound effects to create a highly immersive and interactive environment for language teaching professionals.
Estimation of P(X < Y) using some modifications of ranked set sampling for Weibull distribution
(University of the Punjab, 2017) Akgül, Fatma Gül; Şenoğlu, Birdal
In statistical literature, estimation of R=P(X < Y) is a commonly-investigated problem, and consequently, there have been considerable number of studies dealing with its estimation of it under simple random sampling (SRS). However, in recent years, the ranked set sampling (RSS) method have been widely-used in the estimation of R. In this study, we consider the estimation of R when the distribution of the both stress and strength are Weibull under the modification of RSS, which are extreme ranked set sampling (ERSS), median ranked set sampling (MRSS) and percentile ranked set sampling (PRSS). We obtain the estimators of R using the maximum likelihood (ML) and the modified maximum likelihood (MML) methodologies under these modifications. Then the performances of proposed estimators are compared with the corresponding ML and MML estimators of R using SRS via a Monte-Carlo simulation study.
Optimization of mixture ratios of raw materials in thermoplastic hybrid composites based on particle swarm optimization algorithm
(Springer, 2025) Öztürk, Ercüment; Dönmez Çavdar, Ayfer; Çavdar, Tuğrul
In recent years, the high cost of searching and processing raw materials and the fact that raw material resources have come to the point of depletion force industry and science to seek different solutions. In addition, due to the necessity of increasing the quality, the tendency to composite production has increased. However, it is necessary to make improvements in parameters such as quality, time, and cost in composite production. The use of artificial intelligence technology in the production of composites is becoming more and more common. On the other hand, studies on the determination of raw material mixture ratios are not common. Such studies are usually carried out with experimental productions, which increases the production cost and prolongs the production process. In this study, it was studied on the proportional optimization of the mixture components of thermoplastic hybrid composite materials by using the particle swarm optimization algorithm. First, a dataset was created with the real values obtained from the experimental productions, and the raw material mixture ratios to be included in the mixture were determined by making simulation studies on this dataset. With these obtained ratios, new productions were made, and tests were applied on these productions. Test results have shown that the resulting products are more successful by over 95%. Thus, it has been proven that especially the use of artificial intelligence technology in the field of composites in pre-production stages such as raw material and mixture processes can both reduce the cost and increase the quality.
The digital eye for mammography: deep transfer learning and model ensemble based open-source toolkit for mass detection and classification
(Springer Science and Business Media Deutschland GmbH, 2024) Terzi, Ramazan; Kılıç, Ahmet Enes; Karaahmetoğlu, Gökhan; Özdemir, Okan Bilge
Breast cancer stands as a prevalent malignancy affecting women globally, and a screening method, mammography, boasts reliability for early diagnosis. Nevertheless, interpretive errors during population screening may result in false negatives and positives. To address this, Computer-Aided Detection systems rooted in deep learning have emerged, aiming to reduce both false positive and negative predictions. This study introduces an open-source toolkit called The Digital Eye for Mammography (DEM) and addressing limitations in mammography screening for mass detection and classification. The DEM comprises 11 state-of-the-art object detection architectures and uses a meticulously labeled dataset. It serves as a transfer learning source, and provides ensemble of models from diverse deep-learning architectures, resulting in a more robust solution. Experiments conducted on widely-used datasets indicate that the DEM outperforms existing transfer learning sources by significant margins in terms of true positive rate (TPR). According to the experimental results, the DEM serves as a better transfer learning source for mass detection in pathology-proven InBreast and CBIS-DDSM datasets, presenting improvements 12% and 5% in TPR performance at 0.1 false positive per image (FPPI), respectively. Compared to literature, the DEM achieves lower FPPI values while maintaining higher sensitivity, indicating its potential usage as a transfer learning source. By employing ensemble strategies, the DEM produces more reliable outcomes in our KETEM dataset, reducing FPPI by 49% for BI-RADS 1-2 (Breast Imaging Reporting and Data System) and 46% for BI-RADS 4-5 compared to the best individual model while preserving TPR values. The DEM’s results suggest its ability to attain better performance without requiring complex model hyperparameters optimization. The GitHub repository of the DEM project is publicly available on: https://github.com/ddobvyz/digitaleye-mammography.
A survey on post-quantum based approaches for edge computing security
(John Wiley and Sons Inc, 2024) Karakaya, Aykut; Ulu, Ahmet
With the development of technology and its integration with scientific realities, computer systems continue to evolve as infrastructure. One of the most important obstacles in front of quantum computers with high-speed processing is that its existing systems cause security vulnerabilities. Therefore, in order to take advantage of quantum systems, existing systems that are already secure must also be secure in the post-quantum scenario. One of these systems is edge computing. There are challenges in terms of computational power for the implementation of pre- and post-quantum methods in structures with resource-constrained devices. This article reviews the post-quantum security threats of edge devices and systems and the secure methods developed for them. Although there is relatively little research in this field, it remains relevant. In the studies reviewed, lattice-based approaches are often highlighted for making edge systems quantum-resistant. Additionally, these studies indicate that there has been an increasing trend in this field in recent years. This article is categorized under: Applications of Computational Statistics > Defense and National Security Algorithms and Computational Methods > Networks and Security.
Life fits home: Exploring people's experience with a COVID-19 tracing app in Turkey through a qualitative study
(SAGE Publications Ltd, 2024) Alan, Alper Turan
Mobile apps have been developed to manage COVID-19 in many countries. However, for these apps to be truly effective, they need to be widely adopted by society. To date, there has been less qualitative research on user experiences and perspectives on these apps. The goal of this study was to explore how users perceive and use different features of a COVID-19 tracing app provided in Turkey. Semi-structured interviews (n = 15) were conducted over the phone, audio recorded, and then transcribed verbatim. The transcriptions were analyzed through thematic analysis. The analysis began by categorizing each transcript at the sentence level through open codes, which then grouped into broader themes. In total, 8 male and 7 female participants from 10 different cities took part in the interviews. On average, participants were 32 years (SD = 6.8) old, and their app usage experience was 9.8 months (SD = 3.1). Thematic analysis revealed five key themes: long-term adoption and engagement, perception and desired features, privacy concerns, reliability, and emotional and behavioral impact. The main reasons for users to download and use the app were to check the COVID-19 density map in their region and to access personal health codes for security checks. The COVID-19 density map caused many users to change where they usually travel and shop. Most participants felt comfortable sharing their personal data to collectively manage the pandemic. The majority found the app useful, stating that the app allows them to take precautions against the virus and therefore helps them feel good emotionally. Future contact tracing apps need to provide indicators to enable users to evaluate whether app use is making a difference in the pandemic.
A new parallel tabu search algorithm for the optimization of the maximum vertex weight clique problem
(John Wiley and Sons Ltd, 2024) Dülger, Özcan; Dökeroğlu, Tansel
The efficiency of metaheuristic algorithms depends significantly on the number of fitness value evaluations performed on candidate solutions. In addition to various intelligent techniques used to obtain better results, parallelization of calculations can substantially improve the solutions in cases where the problem is NP-hard and requires many evaluations. This study proposes a new parallel tabu search method for solving the Maximum Vertex Weight Clique Problem (MVWCP) on the Non-Uniform Memory Access (NUMA) architectures using the OpenMP parallel programming paradigm. Achieving scalability in the NUMA architectures presents significant challenges due to the high complexity of their memory systems, which can lead to performance loss. However, our proposed Tabu-NUMA algorithm provides up to (Formula presented.) speed-up with 64 cores for ten basic problem instances in DIMACS-W and BHOSLIB-W benchmarks. And it improves the performance of the serial Multi Neighborhood Tabu Search (MN/TS) algorithm for 38 problem instances in DIMACS-W and BHOSLIB-W benchmarks. We further evaluate our algorithm on larger datasets with thousands of edges and vertices from Network Data Repository benchmark problem instances, and we report significant improvements in terms of speed up. Our results confirm that the Tabu-NUMA algorithm is among the best recent algorithms for solving MVWCP on the NUMA architectures.
Artificial intelligence applications in composites: A survey
(Springer Science and Business Media B.V., 2024) Öztürk, Ercüment; Dönmez Çavdar, Ayfer; Çavdar, Tuğrul
It is known that raw material resources have reached the point of depletion. Therefore, the search for alternative sources is becoming more and more common. The only product that can be considered as an alternative to raw material sources is composites. With the increase in its use in the industrial fields, studies in relation to increasing the quality of composites and reducing the production cost have recently gained attention. Experimental studies based on personal experience have now left their place to Information Technologies. Because IT is a good approach that can provide a solution to the improvement of low quality, long timeframes, and high cost in the experimental studies process. In this context, Artificial Intelligence technologies have the potential to provide better solutions and results. In this survey, a literature review on composites using AI technology was conducted. We have mainly focused on the foundations of the AI technology and its advantages in the field of composites. Consequently, it has been seen that the production of composites via IT approaches increases the quality, reduces the production costs, and abridges the experimental production process.
A parameter-uniform weak Galerkin finite element method for a coupled system of singularly perturbed reaction-diffusion equations
(Faculty of Sciences and Mathematics, University of Nis, Serbia, 2023) Toprakseven, Şuayip; Zhu, Peng
The aim of this paper to investigate a weak Galerkin finite element method (WG-FEM) for solving a system of coupled singularly perturbed reaction-diffusion equations. Each equation in the system has perturbation parameter of different magnitude and thus, the solutions will exhibit two distinct but overlapping boundary layers near each boundary of the domain. The proposed method is applied to the coupled system on Shishkin mesh to solve the problem theoretically and numerically. Elimination of the interior unknowns efficiently from the discrete solution system reduces the degrees of freedom and, thus the number of unknown in the discrete solution is comparable with the standard finite element scheme. The stability and error analysis of the proposed method on the Shishkin mesh are presented. We show that the method convergences of order O(N?k lnk N) in the energy norm, uniformly with respect to the perturbation parameter. Moreover, the optimal convergence rate of O(N?(k+1)) in the L 2 -norm and the superconvergence rate of O((N?2k ln2k N) in the discrete L ?-norm is observed numerically. Finally, some numerical experiments are carried out to verify numerically theory.
3D-CNN and autoencoder-based gas detection in hyperspectral images
(IEEE, 2023) Özdemir, Okan Bilge; Koz, Alper
The detection of gas emission levels is a crucial problem for ecology and human health. Hyperspectral image analysis offers many advantages over traditional gas detection systems with its detection capability from safe distances. Observing that the existing hyperspectral gas detection methods in the thermal range neglect the fact that the captured radiance in the longwave infrared (LWIR) spectrum is better modeled as a mixture of the radiance of background and target gases, we propose a deep learning-based hyperspectral gas detection method in this article, which combines unmixing and classification. The proposed method first converts the radiance data to luminance-temperature data. Then, a 3-D convolutional neural network (CNN) and autoencoder-based network, which is specially designed for unmixing, is applied to the resulting data to acquire abundances and endmembers for each pixel. Finally, the detection is achieved by a three-layer fully connected network to detect the target gases at each pixel based on the extracted endmember spectra and abundance values. The superior performance of the proposed method with respect to the conventional hyperspectral gas detection methods using spectral angle mapper and adaptive cosine estimator is verified with LWIR hyperspectral images including methane and sulfur dioxide gases. In addition, the ablation study with respect to different combinations of the proposed structure including direct classification and unmixing methods has revealed the contribution of the proposed system.
Uphill resampling for particle filter and its implementation on graphics processing unit
(Science Direct, 2023) Dülger, Özcan; Oğuztüzün, Halit; Demirekler, Mübeccel
We introduce a new resampling method, named Uphill, that is free from numerical instability and suitable for parallel implementation on graphics processing unit (GPU). Common resampling algorithms such as Systematic suffer from numerical instability when single precision floating point numbers are used. This is due to cumulative summation over the weights of particles when the weights differ widely or the number of particles is large. The Metropolis and Rejection resampling algorithms do not suffer from numerical instability as they only calculate the ratios of weights pairwise rather than perform collective operations over the weights. They are more suitable for the GPU implementation of the particle filter. However, they undergo non-coalesced global memory access patterns which cause their speed deteriorate rapidly as the number of particles gets large. Uphill also does not suffer from numerical instability but, experiences the same non-coalesced global memory access problem with Metropolis and Rejection. We introduce its faster version named Uphill-Fast which eliminates this problem. We make comparisons of Uphill and Uphill-Fast with the Systematic, Metropolis, Metropolis-C2 and Rejection resampling methods with respect to quality and speed. We also compare them on a highly non-linear system. Uphill-Fast runs faster and attains similar quality, in terms of RMSE, in comparison with Metropolis and Rejection when the number of particles is very large. Uphill-Fast runs with roughly same speed as Metropolis-C2 with better variance and MSE when the number of particles is very large.
DeepSR: a deep learning tool for image super resolution
(ELSEVIER, 2023) Temiz, Hakan
An open source tool is introduced that provides a versatile environment to meet the needs of re- searchers in developing deep learning (DL) algorithms for single image super-resolution reconstruction (SISR). The processes of SISR were carefully studied, unified and integrated to create software that can be used by the community for any type of imaging method such as aerial, medical, optical, etc. DeepSR allows easy implementation of SISR application with rapidly prototyped DL models, and detailed reporting and recording of the results. The entire experiment can be done with simple command line scripts. It can be easily extended by user-defined metrics, augmentations, callbacks, etc.
Error analysis of a weak Galerkin finite element method for two-parameter singularly perturbed differential equations in the energy and balanced norms
(Elsevier, 2023) Toprakseven, Şuayip; Zhu, Peng
A weak Galerkin finite element method is proposed for solving singularly perturbed problems with two parameters. A robust uniform optimal convergence has been proved in the corresponding energy and a stronger balanced norms using piecewise higher order discontinuous polynomials on a piecewise uniform Shishkin mesh. Finally, we give some numerical experiments to support theoretical results.

Güncel Gönderiler