Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7148
Browse
738 results
Search Results
Article Demystifying Power and Performance Variations in Gpu Systems Through Microarchitectural Analysis(Comsis Consortium, 2025) Topcu, Burak; Karabacak, Deniz; Oz, IsilGraphics Processing Units (GPUs) serve efficient parallel execution for general-purpose computations at high-performance computing and embedded systems. While performance concerns guide the main optimization efforts, power issues become significant for energy-efficient and sustainable GPU executions. Profilers and simulators report statistics about the target execution; however, they either present only performance metrics in a coarse kernel function level or lack visualization support that can enable microarchitectural performance analysis or performance-power consumption comparison. Evaluating runtime performance and power consumption dynamically across GPU components enables a comprehensive tradeoff analysis for GPU architects and software developers. In this work, we present a novel memory performance and power monitoring tool for GPU programs, GPPRMon, which performs a systematic metric collection and provides useful visualization views to guide power and performance analysis for target executions. Our simulation-based framework dynamically gathers SM and memory-related microarchitectural metrics by monitoring individual instructions and reports dynamic performance and power values. Our interface presents spatial and temporal views of the execution. While the first demonstrates the performance and power metrics across GPU memory components, the latter shows the corresponding information at the instruction granularity in a timeline. We demonstrate performance and power analysis for memory-bound graph applications and resource-critical embedded programs from GPU benchmark suites. Our case studies reveal potential usages of our tool in memory-bound kernel identification, performance bottleneck analysis of a memory-intensive workload, performance-power evaluation of an embedded application, and the impact of input size on the memory structures of an embedded system.Article Citation - WoS: 1Citation - Scopus: 1Adaptive Reuse of Industrial Heritage in Crises Zones: the Soap Factories in the Levant(Taylor & Francis Ltd, 2025) Saifi, Yara; Zawawi, Zahraa; Yuceer, HulyaThis paper articulates critical considerations for policies related to the adaptive reuse of industrial heritage, particularly in contexts affected by crises, such as conflicts, wars, and natural disasters using the soap factories of the Levant as a case study. It critically investigates the role of adaptive reuse in conserving industrial heritage buildings impacted by crises. The Levant provides a unique context for examining the intersection of cultural heritage and resilience; soap factories, reflective of the region's olive-oil-rich history and industrial legacy dating back to the 18th century, symbolise the conservation challenges and opportunities posed by ongoing regional conflict. The study systematically analyses existing literature on adaptive reuse and questions conventional paradigms in crisis environments, particularly the suitability of standard transformations such as converting industrial heritage into museums or exhibition spaces. Instead, it aims to broaden the dialogue on adaptive reuse by offering context-sensitive policy recommendations. These strategies balance the imperatives of heritage conservation, sustainable development, and contemporary community demands, contributing to a nuanced understanding of industrial heritage preservation in crisis-affected regions.Article Ggnn: Group-Guided Nearest Neighbors for Efficient Image Matching(Springer, 2025) Cine, Ersin; Bastanlar, Yalin; Ozuysal, MustafaThe widely adopted image matching approach remains dependent on exhaustive matching of local features across images. Existing methods aiming to improve efficiency either approximate nearest neighbor (NN) search, compromising accuracy, or apply filtering only after establishing tentative matches, which restricts potential efficiency gains. We challenge the assumption that exhaustive NN search is necessary by proposing a more efficient hierarchical approach that maintains matching accuracy without relying on full-scale NN search. Our key insight is that efficiently identifying sufficiently similar, geometrically meaningful feature matches-rather than the most similar but geometrically random ones-can improve or maintain performance at a lower computational cost. We propose a novel method, Group-Guided Nearest Neighbors (GGNN), which matches groups of features first and then matches individual features only within these matched groups. This hierarchical pipeline reduces the computational complexity of feature matching from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta (n<^>2)$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta (n \sqrt{n})$$\end{document}, significantly improving efficiency. Experimental results on homography estimation demonstrate that GGNN outperforms standard NN search while achieving performance comparable to state-of-the-art methods. Additionally, we formulate GGNN as a general framework, where conventional NN search is a special case with a single global feature group. This formulation provides a continuum of feature matching methods with varying computational costs, enabling automatic selection based on a given time budget.Article Impact of Cooling Strategies and Cell Housing Materials on Lithium-Ion Battery Thermal Management Performance(Mdpi, 2025) Aydin, Sevgi; Çetkin, Erdal; Samancioglu, Umut Ege; Savci, Ismail Hakki; Yigit, Kadri Suleyman; Cetkin, ErdalThe transition to renewable energy sources from fossil fuels requires that the harvested energy be stored because of the intermittent nature of renewable sources. Thus, lithium-ion batteries have become a widely utilized power source in both daily life and industrial applications due to their high power output and long lifetime. In order to ensure the safe operation of these batteries at their desired power and capacities, it is crucial to implement a thermal management system (TMS) that effectively controls battery temperature. In this study, the thermal performance of a 1S14P lithium-ion battery module composed of cylindrical 18650 cells was compared for distinct cases of natural convection (no cooling), forced air convection, and phase change material (PCM) cooling. During the tests, the greatest temperatures were reached at a 2C discharge rate; the maximum module temperature reached was 55.4 degrees C under the natural convection condition, whereas forced air convection and PCM cooling reduced the maximum module temperature to 46.1 degrees C and 52.3 degrees C, respectively. In addition, contacting the battery module with an aluminum mass without using an active cooling element reduced the temperature to 53.4 degrees C. The polyamide battery housing (holder) used in the module limited the cooling performance. Thus, simulations on alternative materials document how the cooling efficiency can be increased.Article Citation - WoS: 1Citation - Scopus: 1Imbalance in Redox Homeostasis Is Associated With Neurodegeneration in the Murine Model of Tay-Sachs Disease(Springer, 2025) Basirli, Hande; Ates, Nurselin; Seyrantepe, VolkanBackgroundTay-Sachs disease is a neurodegenerative disorder characterized by a build-up of GM2 ganglioside in the brain, which results in progressive central nervous system dysfunction. Our group recently generated Hexa-/-Neu3-/- mice, a murine model with neuropathological abnormalities similar to the infantile form of Tay-Sachs disease. Previously, we reported progressive neurodegeneration with neuronal loss in the brain sections of Hexa-/-Neu3-/- mice. However, the relationship between the severity of neurodegeneration and the imbalance in redox homeostasis was not yet clarified in Hexa-/-Neu3-/- mice. Here, we evaluated whether neurodegeneration is associated with oxidative stress in the tissues and cells of Hexa-/-Neu3-/- mice and neuroglia cells from Tay-Sachs patients.Methods and resultsCell death and oxidative stress-related markers were evaluated in four brain regions and fibroblasts of 5-month-old WT, Hexa-/-, Neu3-/-, and Hexa-/-Neu3-/- mice and human neuroglia cells using Western blot, RT-PCR, and immunohistochemistry analyses. We further analyzed oxidative stress levels in the samples using flow cytometry analyses. We discovered neuronal death, alterations in intracellular ROS levels, and damaging effects of oxidative stress, especially in the cerebellum and fibroblasts of Hexa-/-Neu3-/- mice.ConclusionsOur results showed that alteration in redox homeostasis might be related to neurodegeneration in the murine model of Tay-Sachs Disease. These findings suggest that targeting the altered redox balance and increased oxidative stress might be a rational therapeutic approach for alleviating neurodegeneration and treating Tay-Sachs disease.Article Recognition of Counterfactual Statements in Turkish(Assoc Computing Machinery, 2025) Acar, Ali; Tekir, SelmaCounterfactual statements are examples of causal reasoning as they describe events that did not happen and, optionally, those events' consequences if they happened. SemEval-2020 introduces the counterfactual detection (CFD) task and shares an English dataset. Since then, a set of datasets has been released in English, German, and Japanese as part of Amazon product reviews. This work releases the first Turkish corpus of counterfactuals (TRCD). The data collection process is driven by a clue phrase list of counterfactuals, mainly in the form of verb inflections in Turkish. We use clue phrase-based filtering to collect sentences from the Turkish National Corpus (TNC). On the other hand, half of the collection is subject to random word filtering to avoid selection bias due to clue phrases. After the human annotation process with an Inter Annotator Agreement of 0.65, we have 5000 sentences, of which 12.8% contain counterfactual statements. Furthermore, we provide a comprehensive baseline of transformer-based models by testing the effect of clue phrases, cross-lingual performance comparisons using the available CFD datasets, and zero-shot cross-lingual classification experiments using fine-tuning on the different combinations of the existing datasets. The results confirm that TRCD is compatible with the other CFD datasets. Moreover, fine-tuning a Turkish-specific model (BERTurk) performs better than the multilingual alternatives (mBERT and XLM-R). BERTurk is more robust to clue phrase masking. This result emphasizes the importance of a language-specific tokenizer for contextual understanding, especially for low-resource languages. Finally, our qualitative analysis gives insights into errors by different models.Article Virtually Regular Modules(World Scientific Publ Co Pte Ltd, 2025) Buyukasik, Engin; Demir, Ozlem IrmakIn this paper, we call a right module M (strongly) virtually regular if every (finitely generated) cyclic submodule of M is isomorphic to a direct summand of M. M is said to be completely virtually regular if every submodule of M is virtually regular. In this paper, characterizations and some closure properties of the aforementioned modules are given. Several structure results are obtained over commutative rings. In particular, the structures of finitely presented (strongly) virtually regular modules and completely virtually regular modules are fully determined over valuation domains. Namely, for a valuation domain R with the unique nonzero maximal ideal P, we show that finitely presented (strongly) virtually regular modules are free if and only if P is not principal; and that P = Rp is principal if and only if finitely presented virtually regular modules are of the form R-n circle plus (R/Rp)(n)(1) circle plus (R/Rp(2))(n)(2) circle plus center dot center dot center dot circle plus (R/Rp(k))(n)(k) for nonnegative integers n, k, n(1), n(2),...,n(k). Similarly, we prove that P = Rp is principal if and only if finitely presented strongly virtually regular modules are of the form R-n circle plus (R/Rp)(m), where m,n are nonnegative integers. We also obtain that, R admits a nonzero finitely presented completely virtually regular module M if and only if P = Rp is principal. Moreover, for a finitely presented R-module M, we prove that: (i) if R is not a DVR, then M is completely virtually regular if and only if M congruent to( R/Rp)(m); and (ii) if R is a DVR, then M is completely virtually regular if and only if M congruent to R-n circle plus ( R/Rp)(m). Finally, we obtain a characterization of finitely generated virtually regular modules over the ring of integers.Review Citation - WoS: 1Citation - Scopus: 2Organ-On Platforms for Drug Development, Cellular Toxicity Assessment, and Disease Modeling(Tubitak Scientific & Technological Research Council Turkey, 2024) Khurram, Muhammad Maaz; Cinel, Gokturk; Yesil Celiktas, Ozlem; Bedir, ErdalOrgans-on-chips (OoCs) or microphysiological platforms are biomimetic systems engineered to emulate organ structures on microfluidic devices for biomedical research. These microdevices can mimic biological environments that enable cell-cell interactions on a small scale by mimicking 3D in vivo microenvironments outside the body. Thus far, numerous single and multiple OoCs that mimic organs have been developed, and they have emerged as forerunners for drug efficacy and cytotoxicity testing. This review explores OoC platforms to highlight their versatility in studies of drug safety, efficacy, and toxicity. We also reflect on the potential of OoCs to effectively portray disease models for possible novel therapeutics, which is difficult to achieve with traditional 2D in vitro models, providing an essential basis for biologically relevant research.Article Machinability Investigation on Cnc Milling of Recycled Short Carbon Fiber Reinforced Magnesium Matrix Composites(Iop Publishing Ltd, 2024) Atasoy, Sahin; Kandemir, SinanThis study investigates the machinability of magnesium matrix composites reinforced with short carbon fibers, which represent novel materials in the field. AZ91 alloy and its composites containing 2.5 and 5 wt% recycled carbon fiber (rCF) reinforcements were used as workpieces. Face milling was conducted using uncoated carbide cutting tools under dry cutting conditions with varied cutting speeds (480-560-640 m min(-1)) and feed rates (0.65-0.8-0.95 mm min(-1)). The experimental design was based on the Taguchi L-9 (3(3)) orthogonal array. Analysis included cutting forces, surface roughness, wear on cutting inserts, and chip morphology to assess machinability. Taguchi, analysis of variance, and regression methods were employed to analyze cutting force and surface roughness results. Findings indicated satisfactory machinability for AZ91 alloy and comparatively poorer performance for the 5 wt% rCF reinforced composite, with increased reinforcement content correlating with higher cutting force and surface roughness. SEM and EDX analyses revealed significant built-up layer formation on cutting inserts, with predominantly spiral-shaped continuous chips observed in the experiments. Overall, the study affirmed the machinability of the composites and identified suitable cutting parameters for further investigations.Article Secrecy performance of full-duplex space-air integrated networks in the presence of active/passive eavesdropper, and friendly jammer(Wiley, 2024) Buyuksar, Ayse Betul; Erdoğan, Eylem; Altunbas, IbrahimIn this paper, a full-duplex (FD) space-air ground integrated network (SAGIN) system with passive and active eavesdroppers (PE/AE) and a friendly jammer (FJ) is investigated. The shadowing side information (SSI)-based unmanned aerial vehicle relay node (URN) selection strategy is considered to improve signal-to-interference plus noise power ratio (SINR) at the ground destination unit. To quantify the secrecy performance of the considered scenario, outage probability (OP), interception probability (IP), and transmission secrecy outage probability (TSOP) are investigated in the presence of FJ and PE/AE. The results have shown that aerial AE is an important threat since it can severely degrade the OP of the main transmission link. Furthermore, the FJ can decrease the IP of the eavesdropper by causing interference with the cost of power consumption of URNs. Simulations are performed to verify the theoretical findings.