FSM Vakıf Üniversitesi Araştırma ve Akademik Performans Sistemi
DSpace@FSM, FSM Vakıf Üniversitesi’nin bilimsel araştırma ve akademik performansını izleme, analiz etme ve raporlama süreçlerini tek çatı altında buluşturan bütünleşik bilgi sistemidir.
Güncel Gönderiler
Öğe Türü: Öğe , AGFP: A Deep Attention-Guided Framework for DWT-Based Image Steganography(Wiley, 2026) Çevik, Taner; Çevik, Nazife; Paşaoğlu, Ali; Şahin, Fatih; Kiani, Farzad; Ag, Muhammed SaitThis study introduces a novel attention-guided Discrete Wavelet Transform (DWT)-based steganography framework, named Attention-Guided Feature Perturbation (AGFP), which integrates deep visual attention maps with transform-domain embedding to enhance imperceptibility, robustness, and steganalysis resistance. Unlike recent deep-learning-based steganographic systems such as iSCMIS, JARS-Net, and RMSteg, which achieve high visual fidelity but are susceptible to statistical detection, AGFP perturbs only those wavelet coefficients that are identified as perceptually and statistically stable by attention mechanisms extracted from pre-trained CNN models (VGG19, ResNet50, AlexNet, and GoogLeNet). The proposed method is evaluated on the USC-SIPI dataset and the BOSSBase 1.01 benchmark. Experimental results show that AGFP achieves PSNR values between 64.29 and 55.43 dB and SSIM scores between 0.9999 and 0.9989 across varying payloads, indicating consistently high visual quality. While iSCMIS reports slightly higher PSNR and SSIM values, AGFP significantly outperforms all compared methods in bit error rate (BER)—achieving 0.01–0.12, compared to 0.45–0.47 for iSCMIS, 0.31–0.37 for RMSteg, and 0.57–0.75 for JARS-Net. Furthermore, AGFP attains the lowest RS, SPA, and SRM steganalysis detection scores among both classical and deep-learning-based systems. These results confirm that AGFP offers a more balanced and secure steganographic solution, combining high imperceptibility with substantially enhanced robustness and detectability resistance, positioning it as a strong alternative to recent deep-learning-based steganographic frameworks.Öğe Türü: Öğe , Revisiting Workplace Mobbing: Tweets and Qualitative Analysis in Türkiye Case(Springer Nature, 2026) Fındıklı, Mine Afacan; Morgül, Gözde; Anka, Fateme Aysin; Sahmoud, Shaaban; Kiani, FarzadThe globality of mobbing points to huge influence of economic issues over social and societal aspects in the life dynamics of work. COVID-19 presents a new kind of crisis that transforms these factors and establishes new norms in working life simultaneously. Mobbing is to be defined, in this perspective, as the modifications of situation of work and expectations of workers retraining the boundaries and manifestation of mobbing. This study examines the impact of dislocating mobbing, which is a kind of violence that deteriorates the quality of life for employees as well as workplace productivity, in terms of the new dynamics of mobbing and existing dimensions of mobbing-the COVID-19 perspective. Mixed methods research was carried out through macrolevel collection and analysis of tweet data alongside micro-level focus group interviews. While macro findings identified general mobbing dimensions, micro findings revealed more indirect, implicit and specific means of power imbalance. The findings of the research identify emerging gaps in organisational practice regarding diversity and inclusion via the lens of increasing and latent specific power imbalances. In both data analyses, a new dimension of mobbing was identified: the perception of injustice. The emergence of injustice as a new dimension provides a more comprehensive perspective on current practices. The findings of this research are expected to provide valid approaches towards reiteration of existing organisational practices and human resources training.Öğe Türü: Öğe , Bioactive 3D Bioprinted N,S-Graphene Quantum Dot Reinforced Nanocellulose/Fucoidan Scaffolds for Wound Healing(Elsevier, 2026) Çiftçi, Fatih; Sillanpää, MikaThe development of printable bioinks that simultaneously possess superior rheological fidelity and multifunctional bioactivity remains a critical challenge in extrusion-based 3D bioprinting for tissue engineering. Herein, we engineered a novel nanocomposite hydrogel scaffold comprising a structural Cellulose Nanofiber (CNF) backbone and a bioactive Fucoidan (FUC) matrix, reinforced with hydrothermally synthesized Nitrogen and Sulfur co-doped Graphene Quantum Dots (N,S-GQDs). Comprehensive physicochemical characterization confirmed the successful integration of ultrasmall (~9.28 nm), crystalline N,S-GQDs into the polymer network. Rheological analysis revealed that the incorporation of GQDs significantly modulated the viscoelastic properties; all formulations exhibited characteristic non-Newtonian pseudoplastic (shear-thinning) behavior beneficial for extrusion, while the storage modulus (G') consistently dominated the loss modulus (G") across the frequency range, indicating the formation of a stable, solid-like gel structure with enhanced shape fidelity post-printing. Beyond mechanical reinforcement, the nanocomposites demonstrated exceptional biological functionality. The optimized scaffolds exhibited potent, dose-dependent antibacterial activity against Staphylococcus aureus, Escherichia coli, and Pseudomonas aeruginosa, alongside a significant anti-inflammatory efficacy characterized by a 78.4% inhibition of protein denaturation. In vitro biological assessments revealed a transition from passive biocompatibility to active regeneration; the scaffolds induced a remarkable proliferative response in L929 fibroblasts, with cell viability exceeding 140% over 14 days. Furthermore, in a proliferation-independent scratch assay, the GQD-functionalized hydrogels significantly accelerated fibroblast migration, achieving near-complete wound closure (99.8%) within 48 h compared to 55.3% in the control group. These findings collectively establish the 3D printed CNF/FUC/N,S-GQD hydrogels as a robust, rheologically tunable, and bioactive “all-in-one” platform for advanced wound healing strategies.Öğe Türü: Öğe , Marital Problem Solving, Marital Offense Forgiveness, Spousal Self-Efficacy as Predictors of Marital Adjustment(Sage, 2026) Bingöl, Tuğba YılmazMarital adjustment can be conceptualized as a resource of the family system or a part of the regenerative powers of the family. It is important to investigate marital adjustment as it is a concept that deeply affects the quality of life. In this study, which attempts to explain marital adjustment, regression analysis was performed on the obtained results. The data were examined with correlation analysis and regression analysis. According to the study results, there was a strong positive correlation (r = .76) between marital adjustment and spousal self-efficacy among the predictor and dependent variables. Martial adjustment establishes a high level and significant relationship with spouse selfefficacy, resentment-avoidance, marital problem solving, gender, marriage duration, having children and the number of children, income level, education status and place of residence. All variables (spousal self-efficacy, resentment-avoidance, marital problemsolving skills, gender, duration of marriage, having children, number of children, income level, educational status, and place of residence) explain 75% of marital adjustment.Öğe Türü: Öğe , TripletMAML: A Metric-Basedmodel-Agnostic Meta-Learning Algorithm for Few-Shot Classification(Springer Nature, 2026) Gülcü, Ayla; Kuş, Zeki; Özkan, Ismail Taha Samed; Karakuş, Osman FurkanIn this paper, we introduce TripletMAML, a new meta-learning algorithm that enhances the Model-Agnostic Meta-Learning (MAML) approach by incorporating a metric-learning dimension. This enhancement involves the adoption of MAML’s optimization strategies while transitioning to a triplet network model to facilitate metric learning. A novel aspect of this approach is our triplet-task generation technique, designed to produce meta-learning tasks with triplets for both 1-shot and 5-shot settings. TripletMAML extends MAML by jointly incorporating metric-learning and optimization-based principles through a triplet-task formulation, offering a unified and effective framework for few-shot classification.We evaluate Triplet- MAML’s effectiveness across four well-known few-shot image classification benchmarks, comparing its performance against a range of baseline methods. Our findings indicate that TripletMAML, even without data augmentation or extensive hyperparameter adjustments, significantly improves MAML’s performance and surpasses competing baseline approaches in both 1-shot and 5-shot settings. We also demonstrate that optimizing the hyper-parameters automatically using differential evolution method can elevate TripletMAML’s performance to that of more sophisticated models. Additionally, we conduct image retrieval experiments to ascertain whether TripletMAML’s few-shot classification training provides a good starting point for addressing few-shot image retrieval challenges.
