Yıldız Technical University Research Information System
Not a member yet
91324 research outputs found
Sort by
New Generation Antibiotics Derived from DABCO-Based Cationic Polymers
No full textBackground/Objectives: The growing threat of antibiotic resistance necessitates the development of novel antimicrobial agents that effectively target pathogenic microorganisms while minimizing toxicity. Methods: Two series DABCO-based cationic homopolymers (D-subs 1kDa, D-subs 5kDa, D-subs 15kDa) and DABCO–pyridinium-based copolymers (PyH-subs 5kDa_Dsubs 5kDa, PyH-subs 7kDa_Dsubs 3kDa, PyH-subs 3kDa_Dsubs 7kDa) were synthesized to mimic to host-defense cationic peptides via ring-opening metathesis polymerization (ROMP). The antimicrobial activities of these polymers were determined by their minimum inhibitory concentrations (MICs) against E. coli (Gram-negative bacteria), P. aeruginosa (Gram-negative bacteria), S. aureus (Gram-positive bacteria), and C. albicans (fungus). In vitro cytotoxicity assays revealed selective toxicity towards bacterial cells, with high selectivity indices for several copolymers. To gain insight into the mechanism of action, morphological changes in S. aureus upon exposure to D-subs 1kDa were examined using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Results: The D-subs 15kDa homopolymer demonstrated the highest overall antimicrobial activity, particularly against S. aureus (MIC: 8 µg/mL), with all polymers exhibiting minimal hemolytic activity (HC50 ≥ 1024 µg/mL). SEM and TEM results revealed membrane disruption indicative of polymer–bacteria interactions. Additionally, stability studies confirmed polymer integrity under physiological conditions for at least 28 days. Conclusions: These results support the potential of DABCO-based cationic polymers as a promising platform for next-generation antimicrobial therapeutics
OPTICAL SENSOR AND RELAY BASED APPLICATION MODEL FOR ARC FLASH PROTECTION IN MEDIUM VOLTAGE DISTRIBUTION SYSTEMS
No full textAn Efficient Dropout for Robust Deep Neural Networks
No full textOverfitting remains a major difficulty in training deep neural networks, especially when attempting to achieve good generalization in complex classification tasks. Standard dropout is often employed to address this issue; however, its uniform random inactivation of neurons typically leads to instability and insufficient performance increases. This paper proposes an upgraded regularization technique merging adaptive sigmoidal dropout with weight amplification, seeking to dynamically adjust neuron deactivation depending on weight statistics, activation patterns, and neuron history. The proposed dropout process uses a sigmoid function driven by a temperature parameter to determine deactivation likelihood and incorporates a “neuron recovery” step to restore important activations. Simultaneously, the method amplifies high-magnitude weights to select crucial traits during learning. The proposed method is tested on CIFAR-10, and CIFAR-100 datasets using four unique CNN architectures, including deep and residual-based models, to evaluate the approach. Results demonstrate that the suggested technique consistently outperforms both standard dropout and baseline models without dropout, yielding higher validation accuracy and lower, more stable validation loss across all datasets. In particular, it demonstrated superior convergence and generalization performance on challenging datasets such as CIFAR-100. These findings demonstrate the potential of the proposed technique to improve model robustness and training efficiency and provide an alternative in complex classification tasks
Validation of the Short Form of the Remote Work Stress Scale
No full textExtant literature on remote work stress has yet to yield a reliable scale. This study aims to refine the previously established 5-factor, 15-item Remote Work Stress Scale into a unidimensional construct comprising 5 items. As part of the research, we conducted a survey of 602 employees in Turkey who currently actively work remotely. The results showed that the 5 item Short form of the Remote Work Stress Scale is valid (X2/df= 4.91; RMSEA=.08; SRMR=.02; NFI=.99; NNFI=.98; CFI=.99; GFI=.99; AGFI=.95) and reliable (Cronbach’s Alpha=.88; Guttman Split-Half Coefficient=.72). In addition, to examine how the Remote Work Stress Scale differs according to demographic factors, we used multiple correspondence analysis and found that remote work stress is mainly affected by the sex, education and job position. Accordingly, male employees in managerial positions, working in private companies with university or lower education experienced lower remote work stress whereas female and non-managerial employees with master or higher education experienced higher remote work stress.</p
Leaf–like copper–based nanocomposites as adsorbents for dispersive solid-phase extraction: application to determination of cobalt in lemon balm tea using FAAS
No full textThis study proposed an analytical method named dispersive solid phase extraction–flame atomic absorption spectrometry (DSPE-FAAS) for the quantitation of cobalt in balm tea samples. Copper-based nanocomposites were used as adsorbent in the DSPE process. Influential DSPE parameters such as pH/volume of buffer solution, nanocomposite amount, sample volume, mixing type/period, and eluent concentration/volume were optimized to augment signal-to-noise ratio of cobalt. System analytical performance study for the DSPE-FAAS method was carried out, and limit of detection/quantitation (LOD/LOQ) values were recorded as 5.80 µg/kg and 19.33 µg/kg, with a relatively wide dynamic range (20.59 – 400.68 µg/kg). Spiked lemon balm tea samples were employed to perform recovery studies, and satisfactory recovery results were obtained between 76.7% and 128.1% via the external standard calibration method. According to the recorded recovery results, the proposed DSPE-FAAS method can be accurately applied to lemon balm tea samples in order to determine cobalt content
Search for a Neutral Gauge Boson with Nonuniversal Fermion Couplings in Vector Boson Fusion Processes in Proton-Proton Collisions at sqrt[s]=13 TeV
No full textThe first search for a heavy neutral spin-1 gauge boson (Z^{'}) with nonuniversal fermion couplings produced via vector boson fusion processes and decaying to tau leptons or W bosons is presented. The analysis is performed using LHC data at sqrt[s]=13 TeV, collected from 2016 to 2018 with the CMS experiment and corresponding to an integrated luminosity of 138 fb^{-1}. The data are consistent with the standard model predictions. Upper limits are set on the product of the cross section for production of the Z^{'} boson and its branching fraction to ττ or WW. The presence of a Z^{'} boson decaying to τ^{+}τ^{-} (W^{+}W^{-}) is excluded for masses up to 2.45(1.60) TeV, depending on the Z^{'} boson coupling to standard model weak bosons, and assuming a Z^{'}→τ^{+}τ^{-} (W^{+}W^{-}) branching fraction of 50%
Identifying School Travel Mode Choice Patterns in Mersin, Türkiye
No full textThis study investigates the factors affecting the choice of school travel mode among students in Mersin, Türkiye, focusing on walking, private car, public transit and school bus. A two-step modeling approach was adopted. First, a latent class cluster analysis (LCCA) was applied to identify subgroups of students with similar characteristics. Then, separate multinomial logit (MNL) models were estimated for each cluster. The data come from the 2022 Urban Transport Master Plan household survey and include 2798 students from 2092 households. The results show that trip distance is the most consistent and significant factor across all clusters, as increasing distance makes students more likely to use motorized modes instead of walking. Gender also demonstrates a consistent influence in specific clusters, where male students are less likely to travel by private car. Similarly, residing in a single-family house consistently increases the likelihood of car use in multiple clusters. Conversely, the influence of household structure, parental education, income, and household size differs significantly between clusters, underlining the importance of considering group-level differences in school travel behavior. These findings suggest that policies aiming to promote sustainable school travel should be sensitive to the needs of different student groups. Integrating land use and transportation planning may help to support active and shared modes of travel
Mesenchymal Stem Cell-Engrafted Bacterial Cellulose and Graphene Oxide Scaffolds Enhance Peripheral Nerve Repair in a Rat Model
No full textPeripheral nerve injuries result in significant functional impairment, and limited regenerative capacity within the central nervous system further complicates recovery. This study investigates the effects of graphene oxide-decorated bacterial cellulose (BC/GO) scaffolds, with or without mesenchymal stem cells (MSCs), on axonal regeneration following sciatic nerve injury in rats. Twenty-seven male rats were assigned to autograft, BC/GO, and BC/GO+MSCs. The sciatic functional index (SFI), electromyography (EMG), and histopathological analysis were evaluated at 8 weeks. Although SFI scores showed no significant differences, compound muscle action potential (CMAP) values at 4 weeks were significantly higher in both the BC/GO and BC/GO+MSCs groups compared to autografts. Macroscopic examination revealed extensive tissue adhesions in the BC/GO and BC/GO+MSCs groups. Histological analysis indicated regeneration across all groups. The autograft group showed no inflammation, whereas the BC/GO group demonstrated the highest levels of inflammation and degeneration. The BC/GO+MSCs group exhibited reduced inflammation, likely due to the immunomodulatory effects of MSCs. While BC/GO scaffolds promoted early regeneration, the inflammatory response compromised the long-term outcomes. These findings suggest BC/GO scaffolds can facilitate initial nerve repair but require further refinement to sustain long-term functional recovery
Low-Code Platform Selection Using Interval Picture Fuzzy TOPSIS
No full textA low-code platform is a software development environment that enables users to build software applications rapidly with a minimal amount of coding. These platforms provide visual tools, drag-and-drop components, and pre-built logic, allowing users to easily design, develop, and deploy applications. Picture Fuzzy (PF) sets represent experts’ evaluations by using the degrees of membership, non-membership, and hesitancy regarding an element’s inclusion in a set. PF sets have been extended to Interval-Valued Picture Fuzzy Sets (IVPFSs) to offer greater flexibility in assigning these degrees. With this structure, interval-valued picture fuzzy multi-criteria decision-making methods yield more accurate and realistic results compared to other approaches, as they can better capture the opinions of experts. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is a widely used multi-criteria decision-making (MCDM) method for ranking and selecting the best alternative from a set of options based on multiple, often conflicting, criteria. In this study, we propose an evaluation and selection decision model for low-code platforms using the IVPFS-TOPSIS methodology
