AGU GCRIS Premium Database (Abdullah Gül University)
Not a member yet
2706 research outputs found
Sort by
Fabrication and in Vitro Evaluation of a Multifunctional PLA/Gelatin/Cinnamon Membrane for Wound Healing Applications
No full textDeep-Learning AI-Model for Predicting Dental Plaque in the Young Permanent Teeth of Children Aged 8-13 Years
No full textBackground/Objectives: Dental plaque is a significant contributor to various prevalent oral health conditions, including caries, gingivitis, and periodontitis. Consequently, its detection and management are of paramount importance for maintaining oral health. Manual plaque assessment is time-consuming, error-prone, and particularly challenging in uncooperative pediatric patients. These limitations have encouraged researchers to seek faster, more reliable methods. Accordingly, this study aims to develop a deep learning model for detecting and segmenting plaque in young permanent teeth and to evaluate its diagnostic precision. Methods: The dataset comprises 506 dental images from 31 patients aged between 8 and 13 years. Six state-of-the-art models were trained and evaluated using this dataset. The U-Net Transformer model, which yielded the best performance, was further compared against three experienced pediatric dentists for clinical feasibility using 35 randomly selected images from the test set. The clinical trial was registered on under the ID NCT06603233 (1 June 2023). Results: The Intersection over Union (IoU) score of the U-Net Transformer on the test set was measured as 0.7845, and the p-values obtained from the three t-tests conducted for comparison with dentists were found to be below 0.05. Compared with three experienced pediatric dentists, the deep learning model exhibited clinically superior performance in the detection and segmentation of dental plaque in young permanent teeth. Conclusions: This finding highlights the potential of AI-driven technologies in enhancing the accuracy and reliability of dental plaque detection and segmentation in pediatric dentistry
Forecasting the Consumer Price Index in Türkiye Using Machine Learning Models: A Comparative Analysis
No full textThis study utilizes machine learning models to forecast Türkiye's Consumer Price Index (CPI), thereby addressing a critical gap in inflation prediction methodologies. The central research problem involves the forecasting of CPI in a volatile economic environment, which is essential for informed policymaking. The primary objective of this study is to evaluate the performance of three machine learning models, such as Decision Tree (DT), Random Forest (RF), and Support Vector Machine (SVM), in forecasting CPI over periods ranging from one to six months, utilizing data from 2012 to 2024. The study's unique contribution lies in the application of the \"SelectKBest\" method, which identifies the most relevant indices, thereby enhancing the efficiency of the models. An ensemble method, Averaging Voting, is also employed to combine the strengths of these models, producing more accurate and robust predictions. The findings indicate that while the RF model consistently generates the most accurate forecasts across all shifts, the SVM model demonstrates a particular strength in the domain of short-term predictions. The ensemble model demonstrates a substantial performance improvement, with a R2 value of 0.962 for one-month ahead of estimates and 0.956 for five-month forecasts. This combined approach has been shown to outperform individual models, offering a more reliable framework for CPI forecasting. The findings offer valuable insights for economic policymakers, enabling more precise and stable inflation predictions in Türkiye
N-Type Molecular Thermoelectrics Based on Solution-Doped Indenofluorene-Dimalononitrile: Simultaneous Enhancement of Doping Level and Molecular Order
No full textAndreasen, Jens Wenzel/0000-0002-3145-0229; Wang, Suhao/0000-0002-6295-7639The development of n-type organic thermoelectric materials, especially pi-conjugated small molecules, lags far behind their p-type counterparts, due primarily to the scarcity of efficient electron-transporting molecules and the typically low electron affinities of n-type conjugated molecules that leads to inefficient n-doping. Herein, the n-doping of two functionalized (carbonyl vs dicyanovinylene) indenofluorene-based conjugated small molecules, 2,8-bis(5-(2-octyldodecyl)thien-2-yl)indeno[1,2-b]fluorene-6,12-dione (TIFDKT) and 2,2 '-(2,8-bis(3-alkylthiophen-2-yl)indeno[1,2-b]fluorene-6,12-diylidene)dimalononitrile (TIFDMT) are demonstrated, with n-type dopant N-DMBI. While TIFDKT shows decent miscibility with N-DMBI, it can be hardly n-doped owing to its insufficiently low LUMO. On the other hand, TIFDMT, despite a poorer miscibility with N-DMBI, can be efficiently n-doped, reaching a respectable electrical conductivity of 0.16 S cm-1. Electron paramagnetic resonance measurements confirm the efficient n-doping of TIFDMT. Based on density functional theory (DFT) calculations, the LUMO frontier orbital energy of TIFDMT is much lower, and its wave function is more delocalized compared to TIFDKT. Additionally, the polarons are more delocalized in the n-doped TIFDMT. Remarkably, as indicated by the grazing-incidence wide-angle X-ray scattering (GIWAXS), the molecular order for TIFDMT thin-film is enhanced by n-doping, leading to more favorable packing with edge-on orientation and shorter pi-pi stacking distances (from 3.61 to 3.36 & Aring;). This induces more efficient charge transport in the doped state. Upon optimization, a decent thermoelectric power factor of 0.25 mu Wm-1K-2 is achieved for n-doped TIFDMT. This work reveals the effect of carbonyl vs dicyanovinylene on the n-doping efficiency, microstructure evolution upon doping and thermoelectric performance, offering a stepping stone for the future design of efficient n-type thermoelectric molecules. N-doping of two functionalized (carbonyl vs dicyanovinylene) indenofluorene-based conjugated small molecules TIFDKT and TIFDMT is investigated. Remarkably, TIFDMT with a much lower LUMO energy, can be efficiently n-doped to a respectable electrical conductivity of 0.16 S cm-1. Moreover, n-doping of TIFDMT leads to more favorable packing and shorter pi-pi stacking distances, resulting in efficient charge transport in the doped state. imageAgence Nationale de la Recherche [ANR-23-CPJ1-0047-01]; Universit du Littoral Cte d'Opale (ULCO); National Key Research and Development Program [2022YFB3603802]; National Natural Science Foundation of China [62222403, 62074054, U21A20497]; Natural Science Foundation of Hunan Province [2022JJ10019]; Shenzhen Science and Technology Innovation Commission [RCYX20200714114537036]; F.R.S.-FNRS [2.5020.11]; AGU-BAP (Abdullah Gl University-Scientific Research Projects Funding Program) [FYL-2018-115]; National Science Centre, Poland [UMO-2019/33/B/ST3/1550]S.W. and H.W. contributed equally to this work. S.W. gratefully acknowledges Agence Nationale de la Recherche (ANR-23-CPJ1-0047-01) and Universite du Littoral Cpte d'Opale (ULCO) for financial support. Y.H. thanks the National Key Research and Development Program (2022YFB3603802), the National Natural Science Foundation of China (62222403; 62074054; U21A20497), the Natural Science Foundation of Hunan Province (2022JJ10019), and Shenzhen Science and Technology Innovation Commission (RCYX20200714114537036) for financial support. Computational resources were provided by the Consortium des Equipements de Calcul Intensif (CECI) funded by F.R.S.-FNRS under Grant 2.5020.11. J.C. is an FNRS research director. H.U. and I.D. acknowledge support from the AGU-BAP (Abdullah Guel University-Scientific Research Projects Funding Program) (FYL-2018-115). W.P. acknowledges the National Science Centre, Poland, through the grant UMO-2019/33/B/ST3/1550. S.W. gratefully acknowledges Prof. Abdelhak Hadj Sahraoui and Dr. Mathieu Bardoux for their help in setting up the equipment
Shear Strength Prediction for Fiber Reinforced Concrete Beams
No full textDiscrete fibers are often used to increase the tensile and shear strengths of reinforced concrete. Influence of fibers on the behavior of shear critical members is quite significant, therefore, it is crucial to accurately estimate the fiber contribution to ultimate strength. In this study, first a comprehensive database of 446 FRC shear critical beams from 51 different experimental studies is compiled and nonlinear correlation analyses are utilized to identify the key parameters affecting the shear strength. Then, parametric equations are developed to obtain interfacial bond strength of fibers and shear strength of beams with different fiber types, volume fractions, aspect ratios and reinforcement detailing. Shear strengths corresponding to both shear and flexural failures are computed to verify the failure mode. Comparison of the predicted and experimental load carrying capacities indicates the improved accuracy of the prediction equation when compared to the code requirements and existing equations. Due to its applicability to FRC beams with different configurations, reinforcement detailing, fiber types and failure modes, the proposed method is feasible for integration into structural codes as a conservative and practical design approach. © 2025 Elsevier B.V., All rights reserved
Data Impact in Urban Practice: Insights of Kayseri Tram Network
No full textBu tez, Kayseri'deki KayseRay hafif raylı sistemi odağında, veri temelli analiz araçlarının kentsel kaliteyi toplu taşıma altyapısı üzerinden nasıl değerlendirebileceğini ve iyileştirebileceğini araştırır. Çalışma, CBS tabanlı mekânsal analizler, yaya akışı (CFD) simülasyonları ve yolcu verisi görselleştirmelerini bir araya getirerek insan ölçeğinde, bütüncül bir tasarım çerçevesi geliştirmeyi amaçlamaktadır. Analizin merkezinde birbirini tamamlayan üç temel kentsel karakteristik- erişilebilirlik- görünürlük, ve canlılık- bulunur; bu boyutlar, veri odaklı tekniklerle ölçülerek mekânsal ve deneyimsel değerlendirmelerle desteklenir. Seçilen dört istasyon -Düvenönü, Cumhuriyet Meydanı, Hunat Hatun ve Büyükşehir Belediyesi- ayrıntılı biçimde incelenmiş, mekânsal verimsizlikler ile erişim boşlukları tespit edilmiş ve bu eksikleri gidermeye yönelik stratejiler önerilmiştir. Yöntemsel yaklaşım, teknik araçları yerinde gözlemlerle harmanlayarak kentsel hareketlilik sistemlerinin fiziksel çevreyle etkileşimine dair kapsamlı bir bakış sunar. Ayrıca, veri kısıtları ve bağlamsal sınırlamalar ele alınarak yerel altyapı ve kurumsal iş birliğinin önemi vurgulanmıştır. Elde edilen bulgular, yaya akışındaki verimsizlikleri, erişilebilirlik eksikliklerini ve kentsel canlılığı artırma yollarını ortaya koyarak mimarlar ve planlamacılar için uygulanabilir çözümler sunar. Tez, veri temelli yöntemlerin sürdürülebilir, kapsayıcı ve optimize edilmiş kent ortamları yaratmadaki potansiyelini göstererek hem akademik literatüre hem de pratik kentsel tasarım süreçlerine katkıda bulunur; böylece küresel akıllı şehir paradigması ile yerel kentsel gerçeklikler arasında köprü kurar ve BM Sürdürülebilir Kalkınma Hedeflerine uyumlu stratejik bir yol haritası sunar.This thesis investigates how data-driven analysis tools can assess and improve urban quality through public transport infrastructure, focusing on the KayseRay light rail system in Kayseri. The study aims to develop a holistic design framework at the human scale by combining GIS-based spatial analyses, crowd flow (CFD) simulations, and passenger data visualizations. At the core of the analysis are three complementary urban characteristics -accessibility, visibility, and vitality- measured using data-driven techniques and supported by spatial and experiential assessments. The four selected stations -Düvenönü, Cumhuriyet Meydanı, Hunat Hatun, and Büyükşehir Belediyesi- are examined in detail, spatial inefficiencies and access gaps are identified, and strategies to address these deficiencies are proposed. The methodological approach combines technical tools with on-site observations to provide a comprehensive view of the interaction of urban mobility systems with the physical environment. Furthermore, data constraints and contextual limitations are addressed, and the importance of local infrastructure and institutional collaboration is emphasized. The findings provide applicable solutions for architects and planners by revealing inefficiencies in pedestrian flow, accessibility gaps, and ways to increase urban vitality. The thesis contributes to both academic literature and practical urban design processes by demonstrating the potential of data-driven methods in creating sustainable, inclusive, and optimized urban environments; thus, bridging the global smart city paradigm with local urban realities and providing a strategic roadmap aligned with the UN Sustainable Development Goals
Kimlik ve Laiklik Ekseninde Sol Bir Arayış: Barış Partisi
No full textTürkiye’de 1990'lar, siyasal çalkantılar ve toplumsal dönüşümlerin belirginleştiği, ülke tarihinin dönüm noktalarından biri olmuştur. Bu dönem, katliamlar, faili meçhul cinayetler ve siyasi yapının parçalanması gibi travmatik olaylarla şekillenirken, aynı zamanda siyasal İslam’ın yükselişi ve Kürt sorununun dramatik gelişmeleri de toplumsal dinamikleri derinden etkilemiştir. Alevî kökenli vatandaşlar, bu kaotik siyasi ortamdan büyük ölçüde etkilenmiş, özellikle Gazi Mahallesi olayları ve Sivas katliamı gibi trajik hadiseler, Alevî toplumu üzerinde kalıcı izler bırakmıştır. Bu olaylar, Alevîlerde siyasete karşı bir güvensizlik ve korku duygusunun yanı sıra, kendilerini siyasal yapı içinde koruma ve daha etkili bir biçimde var olma ihtiyacını da tetiklemiştir. Alevîlerin, siyasal alanda varlıklarını pekiştirme çabaları, Barış Partisi’nin ortaya çıkışına zemin hazırlamıştır. Barış Partisi, 1990'ların siyasal ortamında Alevî kimliği ve laiklik anlayışının birleşiminden doğan bir siyasi hareket olarak şekillenmiştir. Parti, dönemin toplumsal ve siyasal koşulları ışığında, Alevîlerin özgürlük, kimlik ve laiklik taleplerinin bir arada dile getirilmesi için bir platform sunmuştur. Bu bağlamda Barış Partisi’nin kuruluşu, yalnızca Alevîlerin siyasal alandaki etkinliklerini artırma amacını taşımamış, aynı zamanda Türkiye’nin 1990’lar siyaseti ve Alevî toplumu açısından önemli bir dönemin izlerini taşıyan bir örnek olmuştur. Bu hareketin gelişimi ve toplumsal etkileri, dönemin siyasal atmosferiyle ilişkili olarak Alevîlerin siyasal katılımının nasıl şekillendiğini anlamak açısından önemli bir inceleme alanı oluşturmaktadır
Multi-Method Text Summarization: Evaluating Extractive and BART-Based Approaches on CNN/Daily Mail
No full textWith the exponential growth of digital content, efficient text summarization has become increasingly crucial for managing information overload. This paper presents a comprehensive approach to text summarization using both extractive and abstractive methods, implemented on the CNN/Daily Mail dataset. We leverage pre-trained BART (Bidirectional and AutoRegressive Transformers) models and fine-tuning techniques to generate high-quality summaries. Our approach demonstrates significant improvements, with our best model trained on 287 k samples achieving ROUGE-1 F1 scores of 0.4174, ROUGE-2 F1 scores of 0.1932, and ROUGE-L F1 scores of 0.2910. We provide detailed comparisons between extractive methods and various BART model configurations, analyzing the impact of training dataset size and model architecture on summarization quality. Additionally, we share our implementation through an opensource NLP toolkit to facilitate further research and practical applications in the field. © 2025 Elsevier B.V., All rights reserved
Labyrinthine Microstructures With a High Dipole Moment Boron Complex for Molecular Physically Unclonable Functions
No full textThe design and development of novel molecular-physically unclonable functions (PUFs) with advanced encoding characteristics and ease of fabrication have recently attracted attention in cryptography, secure authentication, and anticounterfeiting. Here, we report the development of a new high dipole-moment small molecule, InIm-BF2, a difluoroborate complex of an indolyl-imine ligand, and the fabrication of unique labyrinthine patterns through a facile two-step thin film process under ambient conditions. The new molecule has a dipolar, coplanar π-backbone and arranges in the solid state with antisymmetric cofacial π-stackings (3.86 Å). These properties, along with short C–H···π contacts (2.74–2.88 Å) and nonclassical C–H···F hydrogen bonds (2.47–2.51 Å) (23.4% and 11.5% of the Hirshfeld surfaces, respectively), drive the formation of amorphous molecular PUF patterns with disordered, short-range interactions. Spin-coating followed by thermal annealing at a moderate temperature produces nanoscopic molecular thin films with intricate labyrinthine patterns. These patterns, characterized by interconnected, irregularly shaped, micron-sized (≈50–100 μm) features, exhibit excellent PUF characteristics, verified through advanced image analysis and computational algorithms. Unlike randomly positioned isolated features in classical binarized keys, the interconnected labyrinthine patterns possess rich entropy and complex features, directly authenticated via deep-learning methodologies. Our work not only demonstrates a facile, promising approach to fabricating unique high-entropy PUF patterns but also provides critical insights into designing advanced molecular materials for next-generation security applications. © 2025 The Authors. Published by American Chemical Societ
Strategic Investment in BIST100: A Machine Learning Approach Using Symbolic Aggregate Approximation Clustering
No full textNalici, Mehmet Eren/0000-0002-7954-6916This study employs the Symbolic Aggregate Approximation (SAX) clustering method to enhance investor decision-making on the Borsa Istanbul (BIST100) by identifying companies exhibiting analogous stock movements. The data from 81 BIST100 companies over a three-year period has been analyzed, with a focus on risk minimization and strategic investment. The SAX method, integrated with a dendrogram, categorizes stocks into sector-based and non-sector-based clusters, providing insights for portfolio optimization. The results demonstrate the effectiveness of the method in identifying relevant stock patterns across sectors, aiding in more informed investment decisions. This approach highlights the need for considering multiple factors in investment strategies, offering a new perspective on stock market analysis with advanced clustering techniques