Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleYalnız künye2026Multi level throttled attention for accurate and efficient classification of retinal diseases in OCT images
Celik, Kemal; Celik, Fatih
The classification of retinal diseases using Optical Coherence Tomography images has gained significant attention in recent years. However, due to the heterogeneity of retinal disorders, achieving high classification accuracy remains challenging. While deep learning has improved the identification of various ophthalmic conditions, accurate and early diagnosis is still critical for effective clinical decision-making. Additionally, reducing computational complexity is essential for building scalable diagnostic systems. To address these challenges, we propose a novel Multi-level Throttled Attention (MLTA) model that first applies attention to extract important features, followed by throttling to filter out redundant or less useful features. This process is applied at multiple stages, operating on both Convolutional Neural Network layers and Transformer architectures, and then fuses them finally. This approach optimizes feature selection by preserving the most informative retinal features while discarding redundant ones from multi-level feature maps. As a result, the model enhances classification performance while reducing computational overhead. We evaluate our proposed model on two benchmark OCT datasets: OCT2017 and OCTID, achieving accuracy scores of 99.59 ± 0.04% and 99.71 ± 0.03%, respectively. Our method outperforms existing CNN-based and Vision Transformer-based models, demonstrating superior accuracy and efficiency through this hierarchical attention.
- MakaleAçık Erişim2026Influencer Authenticity in Marketing: A Systematic Literature Review and Future Research Agenda
Celik, Fatih; Ibrahim, Blend; Polat, Eray; Moulard, Julie; Akbulut, Metin
As influencer marketing scales into a mainstream advertising channel, authenticity has become its defining currency, yet consumers and brands struggle to understand what makes a social media influencer (SMI) genuinely credible and why strategic authenticity backfires. Despite growing scholarly interest, research remains fragmented, leaving practitioners without a unified framework to guide their decisions. Through a systematic literature review (SLR) of 138 peer-reviewed articles (2011-2024), using topic modeling and thematic analysis, we identify the authenticity paradox, the inherent conflict that arises when SMIs attempt to commercialize their genuine self-presentation. This tension also explains why recent evidence reports no overall engagement benefit from perceived authenticity. Seven latent topics are synthesized into four integrated axes: (1) ontological tensions at the technological interface, (2) the commercial transparency and trust paradox, (3) ethical negotiation in identity construction, and (4) strategic consumer responses. Our findings reveal how consumers continuously navigate between SMIs' curated performances and perceived genuineness, with self-determination theory, attribution theory, and parasocial interaction theory dominating current explanations, predominantly within USA-based contexts. Beyond mapping the field, we specify three conceptual patterns, namely authenticity as attributed intent, transparency as coherence rather than concealment, and authenticity as a structurally constrained accomplishment. We then translate the gaps they expose into a structured research agenda, offering actionable insights for brands, managers, and platform designers on authenticity management, influencer selection, partnership governance, and performance evaluation.
- MakaleAçık Erişim2025Optimizing Land Image Classification: A Deep Learning Approach with ResNet-101 on the EuroSAT Dataset
Celik, Fatih; Celik, Kemal
The classification of image is essential to make LULC (Land Use Land Cover) maps. However, the classification of land cover plays a vital role for studying and modernizing the land areas. Recently, deep learning (DL) techniques have achieved outstanding performance in the classification of high-resolution images. Different techniques have been employed in traditional methods to identify LULC due to its complex and ever-changing nature. However, these studies have shown improved outcomes despite some restrictions such as inaccuracies and reduced performance. To address these problems, the proposed study introduces a Squeeze Synchronization Layer (SSL) and a Convolve Craft Focus Module (CCFM) where, SSL reduces input data complexity by removing noise and irrelevant information from images using pooling and convolutional operations also, CCFM enhances feature extraction to improve land classification accuracy. The EUROSAT land image dataset is utilized for the evaluation of the introduced model. Whereas, the dataset comprises of 64x64 images, which are captured by satellite Sentinel-2A in ResNet 101 input layer. Although, a SSL is suggested, and a CCFM is implemented in the convolutional layer for classifying land images. However, the efficiency of the system is evaluated by measuring performance metrics such as recall, F1score, precision, and accuracy values of the proposed system. The accuracy value of the proposed system is 96% of accuracy, 100% of precision, 100% of recall, and 100% of F1-score, signifies the superior efficiency of the proposed model.
- MakaleAçık Erişim2025Kurtun Dam oscillate characterization with landslide possible effect detection using InSAR observations
Celik, Fatih; Sanli, Fusun Balik; Celik, Kemal; Celik, Ayse
Observing and analyzing areas prone to landslides is crucial due to the extensive destructive capacity of landslides. To monitor landslide activities across diverse areas, remote sensing data and image processing methods are employed. This study proposes a method that integrates Multi-temporal Interferometric Synthetic Aperture Radar and advanced temporal tracking technologies to intensify surveillance of the Kurtun Dam and its valley. According to Sentinel-1 A data, the net displacement rates of the dam crest varied between − 12 and 2 mm/year (LOS). InSAR technology surveys the entire dam catchment area, detecting critical infrastructure such as slopes, spillways, and access roads. The crest of the Kurtun Dam experiences oscillatory deformation during filling activities, but the lateral slopes and downstream abutment remain stable. As reservoir water levels rise, the dam structure exhibits downstream deformations, whereas declining water level results in upstream displacements. Monitoring points located the opposite Ozkurtun settlement and within the reservoir area, both facing the dam, exhibited displacement rates of up to -50 mm monthly. In regions susceptible to landslides, the greatest annual displacement rates were observed to be -54 and − 127 mm/year. This study suggests that integrating 3D modeling with InSAR technology into dam monitoring systems can significantly enhance their effectiveness. At Kurtun Dam and other crucial locations within the catchment area, the integrated approach combined with topographic monitoring, ensures reliable landslide detection and analysis. By advancing monitoring technology for embankment dams, this strategy prioritizes safety and structural integrity assessments.