Lisansüstü Eğitim Enstitüsü Tez Koleksiyonu
Bu koleksiyon için kalıcı URI
Güncel Gönderiler
Öğe Fabrication of molecularly imprinted nanoparticle based quartz crystal microbalance (QCM) sensors for angiotensi̇n-II detection from human serum(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Zenger, Okan; Baydemir Peşint, GözdeAngiotensin II (AngII) is a significant effector peptide of the renin-angiotensin system (RAAS) that serves as a growth factor, regulating cellular growth, fibrosis, and apoptosis. AngII plays a critical role in the pathophysiology of several diseases and disorders, including hypertension, cardiovascular diseases, and infections such as influenza and SARS-CoV-2. It is also being explored for its potential role in tumor progression. However, the spontaneous detection of AngII remains challenging due to its low physiological concentration, instability, and the complex nature of biological fluids. Molecular imprinting is a technique that enables the preparation of highly selective synthetic receptors that mimics the natural recognition sites for the molecule of interest. In this study, molecularly imprinted polymers (MIPs) were synthesized in nanoparticle form (~50 nm) and employed as recognition elements in quartz crystal microbalance (QCM) sensor chips for the selective detection of AngII. Following the synthesis and characterization of AngII-imprinted (AngII-MIP) and non-imprinted (NIP) nanoparticles, QCM-based sensors (AngII-MICqcm and NICqcm) were fabricated and used in kinetic binding studies. The AngII-MICqcm sensor demonstrated high sensitivity, with the ability to detect AngII at concentrations as low as 0.25 pg/mL. Selectivity tests using Angiotensin I (AngI) and vasopressin (Vasp) as competing molecules revealed selectivity ratios of 6.09 and 7.44 in favor of AngII, respectively. The calculated imprinting factor (IF) values were 5.58 for AngII, 3.67 for AngI, and 4.50 for Vasp, confirming the superior selectivity of the MIP-based sensor. Upon reusability test, it was found that after the 10th cycle, the QCM chip retained about 96% of its AngII adsorption capacity. Human serum samples were used in studies to detect AngII using the AngII-MICqcm chip. Even in this complex environment, the instrument was able to detect incredibly minute amounts of AngII (1 pg/mL).Öğe Combustion performance analysis of hydrogen-enriched methane in a custom-designed burner(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Erkan, Anıl; Tüccar, GökhanDue to the increasing global energy demand and growing environmental concerns, efforts have been made to achieve optimal combustion conditions by blending hydrogen gas with conventional fuels at various ratios, taking into account the adverse combustion effects of pure hydrogen. The aim of this study is to investigate the effects of hydrogen addition—at varying concentrations and without premixing—on the flame characteristics of methane combustion within a specially designed burner. The analysis was conducted using ANSYS Fluent computational fluid dynamics (CFD) simulations, and the results were further interpreted and predicted through the application of the artificial neural network (ANN) method. In this study, combustion reactions were performed for 13 different hydrogen concentrations ranging from 0% to 30% in increments of 2.5%, blended with CH4 gas. The simulation results revealed that increasing the hydrogen concentration within the mixture led to higher flame temperatures and generally more stable flame structures. The maximum flame temperature of 2303.82 K was achieved at 30% hydrogen addition. At this concentration, the trend plateaued, indicating rich mixture behavior and a reduced rate of increase. The axial flame profile also became more uniform, with consistent distribution across the burner center. Additionally, maximum velocity values were observed to increase in correlation with rising maximum flame temperatures. As a result of the prediction analyses conducted using ANN and regression equations, the accuracy of the results was calculated, and the reliability of the CFD studies was evaluated. CFD analyses were validated by comparison with various experimental studies, and the results were found to be in agreement.Öğe Estimation of parameters affecting water quality using data mining algorithms(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Kavak, Elif; Kaya Keleş, MümineAchieving a sustainable life is one of the most important issues today. In order for future generations to achieve this life, the United Nations (UN) has published the Sustainable Development Goals (SDGs). Among these, the 6th SDG under the title of "Clean Water and Sanitation" is of critical importance in terms of human health, hygiene and protection of ecosystems. Therefore, it reveals the necessity of sustainable management and effective monitoring of water resources. In this context, data mining applications and regression models were developed in the study carried out in order to monitor water quality and predict future changes. Within the scope of the study, five separate data sets were created by bringing together the parameters affecting water quality obtained from drinking water analyses of Adana, Mersin, İzmir, Sakarya and İstanbul provinces of Turkey. Water Quality Index (WQI) was calculated using various physical and chemical parameters. Various algorithms such as Linear Regression (LR), Support Vector Regression (SVR), Artificial Neural Networks (ANN), Decision Tree (DT), Random Forest (RF), Ridge Regression (RR) and Lasso were evaluated comparatively. As a result of the study, it was seen that ANN, SVR and LR models were effective for water quality management among the models evaluated with two different evaluation metrics.Öğe Dynamic behavior of hemp based composite structures(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Çetin, Aykut; Kurtaran, HasanThis study investigated the free vibration behavior of hemp-based laminated composite structures through both numerical modeling and experimental investigation. The study includes two primary structural forms: curved composite beams and laminated rectangular plates. The main objective of the study was to assess the dynamic performance of pure and hybrid hemp fiber-reinforced composites and to evaluate the viability of hemp fibers as a sustainable alternative to conventional synthetic reinforcements, particularly glass fiber. The governing equations for the free vibration of curved beams and laminated plates were derived using the principle of virtual work within the framework of First-Order Shear Deformation Theory (FSDT). For laminated plates exhibiting large-amplitude vibrations, the equations were formulated based on nonlinear Green-Lagrange strain measures. Spatial derivatives in the vibration equations were computed using the Generalized Differential Quadrature (GDQ) method. To determine the mechanical properties, tensile tests were conducted on laminates reinforced with hemp, carbon, and glass fibers. Tensile and vibration test specimens were manufactured using the Vacuum-Assisted Resin Transfer Molding (VARTM) technique. First-mode natural frequencies and damping ratios were measured experimentally using a vibration test setup for nine different composite beams. The developed numerical models were validated through experimental testing and comparison with existing literature, showing a high degree of agreement and confirming the models' accuracy and reliability. After validation, a series of parametric studies was conducted to investigate the influence of curvature ratio, aspect ratio, stacking sequence, boundary conditions, and hybrid fiber configurations on the natural frequencies. The experimental and numerical findings revealed that, in terms of natural frequency performance, the pure carbon composite indicated the highest value. After pure carbon composites, a certain configuration of the carbon/hemp hybrid composites exhibited the highest natural frequency values, higher than the carbon/glass counterpart. CHHHC (C: carbon, H: hemp) and CHCHC stacking sequences consistently indicated higher frequency values than carbon/glass hybrid composites in various configurations, regardless of the boundary conditions. Overall, the results confirm that carefully configured hemp-based hybrid laminates offer promising frequency values for structural applications.Öğe Sample selection for engagement-related EEG recordings(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Arslan, Mustafa Turan; Yıldırım, EsenTraditional approaches to improving EEG classification performance have predominantly focused on feature selection techniques. However, alongside selecting appropriate features, the presence of relevant samples within the dataset plays an equally critical role in boosting classification success. This thesis proposes a novel approach for sample selection from EEG data. To facilitate a robust evaluation of the proposed method, a task-specific EEG dataset was constructed, and the performance of the developed sample selection algorithm was systematically assessed within this framework. EEG dataset was collected from 20 healthy university students and graduates (15 males, 5 females) residing in Adana and Hatay, Turkey. Following data collection, a Continuous Wavelet Transform (CWT) analysis was performed, resulting in two sets of data: Feature-Data (FD) and Task-Engagement-Index (TEI) data, which were subsequently used for classification. To apply sample selection to the dataset, we employed the newly proposed Relief-based Sample Selection (RbSS) algorithm. Primarily, k-Nearest Neighbor (kNN) classification on the raw Feature-Data (FD) without sample selection achieved an accuracy rate of 79.20%. Applying the RbSS algorithm to the FD data demonstrably increased classification accuracy to 79.74%. To improve the classification of Feature-Data (FD), a subsequent sample selection study focused on using TEI data. This two-stage approach involved an initial selection of samples based on the TEI data, followed by the selection of corresponding FD samples based on the index values of the selected TEI data samples. This strategy, incorporating TEI data, yielded a higher classification accuracy of 80.26%.Öğe Performance analysis of VPN protocols over lag architectures under varying MTU conditions(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Kırmızı, Sefa; Demirdelen, TuğçeThis thesis presents a performance comparison of two widely used VPN protocols, WireGuard and IPsec, under varying Maximum Transmission Unit (MTU) sizes and Link Aggregation (LAG) architectures. Various configurations—including Round-Robin, 802.3ad, single interface/single tunnel, and dual interface/dual tunnel setups—were evaluated using key performance metrics such as throughput, packets per second (PPS), CPU utilization, and transmission efficiency. The experiments were conducted on a multi-core Intel-based system using iPerf3 with both single and 20 parallel session tests across five different MTU values (ranging from 450 to 9200 bytes). The results show that WireGuard consistently delivers higher performance and lower CPU usage than IPsec, particularly with larger MTU sizes. While the Round-Robin mode achieved higher PPS, it also increased CPU load and caused packet reordering issues. In contrast, 802.3ad provided more stable multi-core utilization, though with slightly reduced throughput. The most efficient outcomes were observed with dual interface setups using WireGuard and high MTU values. Additionally, a mathematical model was developed to describe the relationship between throughput, MTU, CPU load, and encryption overhead. This model offers a practical tool for network engineers to design optimized VPN architectures, particularly for resource-constrained endpoint devices.Öğe Swarm intelligence algorithms for combinatorial optimization: Encoding and decoding strategies(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Aktaş, Muhammet; Kılıç, FatihThis thesis introduces a new Discrete StarFish Optimization Algorithm (D-SFOA) to solve a complex discrete Symmetric Travelling Salesman Problem (STSP). In the discrete SFOA algorithm, the continuous values of individuals in the population are converted to the discrete version using the random key method. Ten neighbourhood methods used in this study provide diversity to the starfish population, and the 2-opt local search algorithm allows the study to find shorter tours. The performance of D-SFOA is tested on STSP datasets ranging in size from 30 to 1084 from TSPLIB. This thesis also introduces a Modified Choice Function (MCF) to the Discrete Artificial Bee Colony algorithm for adaptive neighbor selection in the symmetric traveling salesman problem. The Metropolis Acceptance Criteria (MAC) are then utilized to provide a chance for more inferior optimal solutions. Discrete versions of the Grey Wolf Optimizer (D-GWO) and Harris Hawk Optimization (D-HHO) algorithms are applied with the same parameters to compare the performance of the proposed algorithms. The algorithm uses descriptive statistics such as average tour, best tour, percentage of deviation of the mean tour, percentage of deviation of the best tour, and execution time to ensure a fair comparison. The Wilcoxon signed-rank test and Ablation test are applied to measure the significant difference in the values of the algorithms and to observe the performance effect of the main components used in the proposed algorithm on tour length and execution time, respectively. The proposed D-SFOA and D-ABC algorithms outperform the other algorithms.Öğe The effect of partial flexibility on aerodynamic performance of finite wing(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi , 2025) Emirhan Eroğlu, Ali; Durhasan, TahirIn this thesis, the aerodynamic effects of partially flexible surface coverage applied to a NACA 0018 airfoil operating in low Re regimes were experimentally investigated. The flexible surface, made of latex material, was placed along a specific chordwise region of the wing, and four different wings with varying spanwise flexibility ratios were tested. Wind tunnel experiments were conducted in the Re range of 3 x 10⁴ to 10 x 10⁴, measuring lift (CL) and drag (CD) coefficients. Additionally, surface oil and tuft flow visualization techniques were used to analyze flow structures. The findings revealed that the partially flexible surface significantly suppressed the laminar separation bubble (LSB), delayed stall, and enhanced lift coefficient, thereby improving aerodynamic efficiency, particularly at lower Re. However, the effectiveness of the flexible surface diminished as the Re increased. These results contribute to the literature by demonstrating the potential of partial flexibility in optimizing passive flow control methods for UAV applications.Öğe Uluslararası örgütler ve doğal afetlerde iletişim: Japonya ve Türkiye örneği(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Büyükce, Ertuğrul; Kılıçbeyli, Elife HatunDoğal afetler, bireylerin yaşamlarını derinden etkileyen, toplumsal yapı üzerinde ekonomik, sosyal ve psikolojik yıkımlara neden olan olaylardır. Bu çalışmada, küresel ısınma gibi doğal afetlerin uluslar ve toplumlar üzerindeki etkileriyle, toplumsal iletişimin ülke koordinasyonundaki önemi incelenmiştir. Afet yönetiminde iletişim eksikliklerinin neden olduğu sorunlar ele alınmış, bilgi akışındaki aksamaların kurtarma ve müdahale süreçlerini nasıl olumsuz etkilediği analiz edilmiştir. Ayrıca, kriz dönemlerinde doğru, hızlı ve güvenilir bilgi paylaşımının afet zararlarını azaltmadaki rolü vurgulanmıştır. Araştırma sonucunda, afet yönetiminde iletişim altyapısının güçlendirilmesi, kurumlar arası koordinasyonun artırılması ve toplumun afet bilincinin geliştirilmesi gibi önerilere yer verilmiştir. Bu bağlamda, etkili bir iletişim stratejisinin, toplum dayanıklılığını artırarak afetlerin zararlarını önemli ölçüde azaltabileceği sonucuna ulaşılmıştır.Öğe Preparation of molecularly imprinted nanoparticle-based surface plasmon resonance (SPR) sensors for angiotensin-II detection from human serum(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Cemek Şahin, Kardelen; Baydemir Peşint, GözdeAngiotensin II (AgII) is a short-chain peptide involved in the functioning of the renin-angiotensin-aldosterone (RAAS) system. It is actively involved in controlling many physiological activities within the human body. It acts as a powerful vasoconstrictor. It also contributes to the maintenance of fluid-electrolyte balance. The importance of AgII is not limited to its physiological regulatory functions; it is also evaluated as a biomarker in various cardiovascular and metabolic diseases. In clinical conditions such as hypertension, heart failure and chronic kidney disease, the observation of increased AgII levels in plasma or urine provides important clues about the underlying physiological processes of these diseases. In this context, measuring AgII levels provides valuable information for diagnosing these diseases in the early stages and monitoring the course of the disease. This study aims to develop molecularly imprinted nanoparticle-based surface plasmon resonance (SPR) sensors for the detection of AgII from human serum. To achieve this, AgII-imprinted (AgII-MIP) and non-imprinted nanoparticles (NIP) were synthesized using the molecular imprinting method. The synthesized nanoparticles underwent detailed analysis through Zeta-size measurements, Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), and FTIR spectroscopy techniques. According to the data obtained from Zeta-size measurements, the particle sizes of AgII-MIP and NIP nanoparticles were determined as 46.51 nm and 48.97 nm, respectively. The nanoparticles were then used to coat the surface of the SPR chip. Thus, SPR chips coated with AgII-MIP nanoparticles (AgII-MICspr) and NIP nanoparticles (NICspr) were successfully obtained. Characterization studies of the chips were carried out by Contact Angle and Atomic Force Microscopy. AgII solutions were made at a range of concentrations (3-100 pg/mL) and these solutions were interacted with the AgII-MICspr chip. The AgII-MICspr chip showed sensitive detection ability even for very low AgII concentration (3 pg/mL). In addition, selectivity studies were performed using AgII-MICspr and NICspr chips. Here, Angiotensin I (AgI) and vasopressin (Vp) were used as competitors. It was determined that the selectivity of the AgII-MICspr chip against AgII was 4.75 and 5.43 times higher than AgI and Vp molecules, respectively. Reusability studies were performed using the AgII-MICspr chip and it was observed that the chip preserved its functionality in the binding regions even after 10 repetitions. AgII detection studies were conducted from human serum. The chip was able to selectively detect very low concentrations of AgII even in this complex environment.Öğe The relationship between financial reporting quality and sustainability reporting disclosures: An application on Borsa Istanbul and German Stock Exchange companies(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Karadağ, Meral; Yaşar, AlpaslanThis thesis empirically examines the relationship between financial reporting quality (FRQ) and sustainability reporting disclosures based on the samples of Turkey and Germany. The study covers data from 75 Turkish companies listed on Borsa Istanbul (BIST) and 140 companies listed on German Stock Exchange (GSE) for the period 2019–2022. All financial and environmental, social and governance (ESG) data employed in this study were sourced from the Refinitiv database, a subsidiary of the London Stock Exchange Group (LSEG). The empirical analysis utilised a range of statistical techniques, including panel data regression analyses, independent samples t-test, and cross-sectional dependence tests. The measurement of financial reporting quality (FRQ) was grounded in earnings management (EM) proxies, specifically the Modified Jones Model and the Performance-Matched Modified Jones Model. In the analyses conducted for the Turkish sample, no statistically significant relationship was identified between overall ESG scores and financial reporting quality. However, the management score variable exhibited a positive and statistically significant association with financial reporting quality as measured by both the Modified Jones Model (FRQ1) and the Performance-Matched Modified Jones Model (FRQ2). Furthermore, within the framework of the Modified Jones Model (FRQ1), a negative and statistically significant relationship was observed between financial reporting quality and the emissions score variable. These findings indicate that certain ESG subcomponents, albeit to a limited extent, exert significant effects on financial reporting quality. Moreover, during the 2019–2021 period, firms engaged in sustainability reporting demonstrated statistically significantly higher financial reporting quality compared to non-reporting firms. However, this difference lost its statistical significance as of 2022. In the analyses for the German sample, financial reporting quality, as measured by the Modified Jones Model (FRQ1) and the Performance-Matched Modified Jones Model (FRQ2), was found to have negative and statistically significant associations with the management score, CSR (Corporate Social Responsibility score) strategy score and emissions trading participation variables. In contrast, a positive and statistically significant relationship was identified between financial reporting quality, under both models, and the governance score. These findings suggest that the variables in question significantly influence the quality of firms' financial reporting practices. The findings of this study demonstrate that the relationship between sustainability reporting and financial reporting quality is influenced not only by institutional contexts but also by external factors such as regulatory frameworks and investor expectations. This outcome underscores the necessity, from the perspective of policymakers and regulatory bodies, of developing strategies aimed at enhancing the content of sustainability disclosures. Furthermore, it provides a meaningful contribution to the international literature based on comparative analyses. Keywords: Borsa Istanbul, ESG Scores, German Stock Exchange, Financial Reporting Quality, Sustainability Reporting Disclosures.Öğe Determination of the effectiveness of artificial intelligence models in detecting maize leaf diseases(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Gökten, Adnan; Tekeli, ErkutMaize is a strategic agricultural product that is of great importance as a source of food and feed; however, leaf diseases cause significant losses in production. The time-consuming nature and high error rate of traditional diagnostic methods have accelerated the search for innovative solutions in agriculture. In this study, the automatic diagnosis of Maize leaf diseases using Convolutional Neural Networks (CNN) was investigated. The dataset obtained from Kaggle, containing images of diseased and healthy leaves, was divided into 80% training, 10% validation, and 10% test sets to compare the performance of different CNN models. The ConvNeXt model demonstrated the highest performance with a 96% accuracy rate, followed by DenseNet with 95% and EfficientNet with 94% accuracy rates. MobileNet stood out with a 92% accuracy rate and low computational cost. The results show that modern CNN architectures provide higher accuracy and efficiency compared to older models. The use of deep learning technologies in agricultural applications holds significant potential in the agricultural sector by offering effective and reliable solutions for disease diagnosis.Öğe Pancar kvass üretiminde farklı laktik asit bakterilerinin fenolik, aroma-aktif bileşikler ve diğer kalite parametreleri üzerine etkileri(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Sevindik, Onur; Kelebek, HaşimBu çalışmada, farklı laktik asit bakterisi (LAB) suşlarının kırmızı pancar (Beta vulgaris var. conditiva) fermantasyon süresi, elde edilen "pancar kvass" ürününün antioksidan aktivitesi, toplam fenolik içeriği, renk özellikleri, mikrobiyal florası, fenolik bileşik profili, aroma, aroma-aktif bileşenleri, duyusal özellikleri ve mineral element içeriği üzerindeki etkileri ilk kez araştırılmıştır. Çalışma kapsamında spontan fermantasyona bırakılan kontrol örneki (K) ve pancar kvass kurulumu aşamasında %1'lik konsantrasyonda suş ekimleri yapılarak Lactobacillus casei (ATCC 431), Lactobacillus delbrueckii (ATCC 9649) ve Lactobacillus plantarum (ATCC 14917) suşlarıyla fermente edilerek üretilen (sırasıyla LC, LD ve LP kodlu) örnekler ile çalışılmıştır. Örneklerde antioksidan kapasite DPPH ve ABTS yöntemleriyle; toplam fenolik madde (TFM) miktarı ise Folin-Ciocalteu yöntemiyle belirlenmiştir. Bulgular, özellikle LP suşunun ilavesinin özellikle antioksidan kapasite üzerinde, LP ve LD suşlarının TFM miktarı üzerinde anlamlı ve pozitif etkiler yarattığını göstermiştir. Pancar kvass örneklerinde yer alan betasiyaninler, betaksantinler ve diğer betalain grubu bileşikler LC-DAD-ESI-MS/MS cihazı ile tanımlanmış ve konsantrasyonları hesaplanmıştır. Toplam 15 fenolik bileşiğin tespit edildiği çalışmada, TFM bulgularıyla paralel şekilde, LAB ilavesi yapılan örneklerin fenolik profilleri kontrol grubuna kıyasla önemli düzeyde değişmiştir. Özellikle L. plantarum ile fermantasyonda toplam betasiyanin içeriği yaklaşık dört kat, betaksantin içeriği ise yaklaşık iki kat artış göstermiştir. Örneklerin aroma profilleri SAFE (Solvent Assisted Flavor Evaporation) yöntemiyle belirlenmiş, aroma-aktif bileşikler ise AEDA (Aroma Extract Dilution Analysis) tekniği ve GC-MS/Olfaktometri ile analiz edilmiştir. Tıpkı fenolik gruplarda olduğu gibi, aroma ve aroma-aktif bileşen profilleri de LAB destekli fermantasyonlarda anlamlı düzeyde farklılıklar göstermiştir. Toplamda 46 aroma bileşiği tanımlanmış; LP ve LD örneklerinde aroma bileşiklerinin toplam konsantrasyonları, K ve LC örneklerine kıyasla anlamlı düzeyde yüksek bulunmuştur. Tüm örneklerde ketonlar ve karboksilik asitler baskın aroma grupları olarak tespit edilmiştir. Ayrıca, GC-MS/O analiziyle toplam 20 aroma-aktif bileşik tanımlanmıştır. Duyusal analizler tanımlayıcı test yöntemiyle gerçekleştirilmiştir. Dört farklı ürün arasında LP örneği, panelistler tarafından aroma, tat ve genel beğeni açısından en yüksek puanı almıştır. Genel değerlendirme sonucunda, kırmızı pancar kvass üretiminde farklı LAB suşlarının kullanılması, fermantasyon süresi, renk, antioksidan kapasite, fenolik, aroma ve aroma-aktif bileşikler, duyusal kalite ve mineral madde içeriği üzerinde belirgin ve olumlu etkiler yaratmış; spontan fermantasyona kıyasla daha üstün kalite özelliklerine sahip ürünlerin elde edilmesini sağlamıştır.Öğe The effect of inclusive leadership on job engagement, organizational trust and innovative behavior in search and rescue units(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Seğmenoğlu, MehmetSearch and Rescue (SAR) is a service whose standards are set by international organizations and is offered free of charge to individuals in danger, regardless of their nationality. SAR operations are sometimes conducted in small areas and sometimes in very large ones, depending on the magnitude of the disaster. While a few SAR units are sufficient for operations conducted in small areas, national and even international units work together in larger areas. Therefore, the unit leader assigned to ensure coordination among personnel from different organizations and to oversee assigned tasks and implemented procedures is considered extremely important in SAR operations. In this context, this study explores the relationship between the concepts of Inclusive Leadership, Job Engagement, Organizational Trust, and Innovative Behavior in the management of new units formed with personnel and volunteers from different organizations and structures. In the study, firstly, a detailed literature review was conducted on the concepts of Inclusive Leadership, Job Engagement, Organizational Trust and Innovative Behavior, then the purpose of the study and the research model were created and the hypotheses to be tested between the concepts were established. In order to test the validity and reliability of the survey prepared within the scope of the study, it was decided to conduct a pilot study. In the study where the cross-sectional method was used in the data collection process, simple random sampling method was applied and volunteering was taken as a basis, and the survey form prepared via googledocs was distributed through social platforms after the necessary permissions were obtained. As a result of the analysis of 118 valid surveys obtained within the scope of the pilot study, it was concluded that the Cronbach's Alpha values of all scales were above the accepted value of .70 for determining reliability (IL: .974; JE: .919; OT: .812; IB: .921), and it was decided to use the scale as in the pilot study. During the main study process, which employed the same methods and procedures as the pilot study, 386 valid surveys were reached, and Exploratory Factor Analysis, Confirmatory Factor Analysis, Correlation Analysis and Hayes Process Macro analyses were conducted with SPSS and AMOS Statistical Programs to test the hypotheses and to determine mediation and serial mediation effects. When the relationships between the variables were examined, it was seen that all the hypotheses established within the scope of the research model were supported, the model developed within the research was verified and it was also determined that Innovative Behavior and Organizational Trust had separate and serial mediating effects between Inclusive Leadership and Job Engagement.Öğe Investigation of the performance of deep learning-based object detection systems(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Dinç, Barış; Kaya, YasinIn recent years, significant advancements have been achieved in object detection (OD) within computer vision due to rapid developments in deep learning (DL) methodologies. Nevertheless, region-based convolutional neural networks (R-CNNs) often suffer from computational inefficiencies due to the generation of an excessive number of candidate regions, many of which are redundant or irrelevant. Although single-stage algorithms have accelerated OD, the presence of high-frequency noise and irrelevant details makes these models sensitive to background disturbances. To adress these limitations, this study introduces a novel approach reducing the high-frequency noise in the input image for the object detection task. Specifically, the proposed model comprises a novel Adaptive Padding (AP) mechanism and Region of Interest (RoI) detector, which provides a balance between RoI generation and object detection. Experimental evaluations conducted on five bird datasets using R-CNN demonstrate that our method increased the proportion of region proposals with an Intersection over Union (IoU) greater than 0,5 from 20,93% to 65,17%. Furthermore, the number of positive proposals increased by 330% during training and 726% during testing, while redundant proposals were reduced by 55,48%. The proposed model was also used for YOLOv8 and tested with 351 images of 6 bird species. The model increased the mAp50 from 0,954 to 0,984 and the mAP50-95 from 0,633 to 0,781.Öğe Machine learning and systems biology-based approaches for identification of key biomolecules in liver diseases(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Cömertpay, Betül; Göv, EsraLiver diseases remain a significant global health concern, causing approximately two million deaths annually. In this thesis, presents a comprehensive systems biology investigation into the molecular landscape of liver diseases by integrating bulk and single-cell transcriptomic, proteomic, regulatory, and microbiome data. Spanning a wide range of clinical conditions including early chronic liver disease (eCLD), chronic liver disease (CLD), acute liver failure (ALF), acute-on-chronic liver failure (ACLF), non-alcoholic fatty liver disease (NAFLD), steatohepatitis, cirrhosis, and hepatocellular carcinoma (HCC) the study identifies both known and novel biomolecular signatures with potential diagnostic, prognostic, and therapeutic relevance. Differential expression analysis, gene regulatory network inference, and single-cell-informed modeling were combined with microbiome profiling to characterize stage-specific molecular changes. Novel candidate biomolecules included genes (e.g., PRKAB2, CLEC1B, CRHBP, TJP2), microRNAs (e.g., miR-548az-5p, miR-548t-5p), and transcription factors (e.g., FOXO1, SP1, ESR2). Additionally, microbial species such as Eikenella and Dictyostelium discoideum emerged as potentially new contributors within the gut-liver axis. Machine learning played a central role in the analytical framework, enhancing both the diagnostic and prognostic interpretability of the results. Diagnostic classification was performed using principal component analysis (PCA) and several supervised algorithms, including Random Forest, k-Nearest Neighbors, Multilayer Perceptron (MLP), XGBoost, LightGBM, Gradient Boosting, and CatBoost. Notably, CatBoost achieved the highest accuracy in HCC diagnosis (93.75%) with perfect sensitivity and specificity (AUC = 1.0). Prognostic modeling using Random Survival Forests (RSF) enabled the stratification of HCC patients by risk level and highlighted key survival-associated genes in specific immune cell types—for instance, RTN3, ENO1, and FOLR3 in monocytes, and CRIP1, PSMA7 in NK cells. Robustness was ensured through 1,000-iteration bootstrap resampling and one-sample t-tests against random baseline models, confirming the reliability of the computational predictions. Altogether, this thesis demonstrates the power of integrating multi-omics data with machine learning and systems-level approaches to reveal critical insights into liver disease biology. The findings not only provide a curated resource of biomolecular signatures across disease states and cell types but also highlight candidates with high translational potential. By combining data-driven discovery with clinical relevance, this work lays a strong foundation for future biomarker validation and precision medicine strategies in liver disease research.Öğe The effect of leadıng edge morphıng on the aerodynamıc characterıstıcs of naca 0012 aırfoıl at low reynolds number(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Özdemiroğlu, Tolga Baran; Durhasan, TahirIn this study, it is aimed to enhance the aerodynamic performance of the NACA 0012 airfoil with the morphing leading edge method. The investigation was conducted using an airfoil with a chord length of 120 mm under Reynolds number conditions of Re = 105. In addition to the base airfoil, a MATLAB code was developed to generate modified airfoil geometries by specifying the deflection angle and the deflection initiation point (expressed as a percentage of the chord length) as input parameters. The computational analysis considered deflection angles of 2°, 4°, and 6°, initiated at 10%, 20%, and 30% of the chord length. All configurations were analyzed using ANSYS FLUENT and compared aerodynamically with the base case (0° deflection). The performance evaluation focused on maximizing lift coefficient (CL), minimizing drag coefficient (CD), and optimizing the lift-to-drag ratio (CL/CD) as a measure of aerodynamic efficiency. The results demonstrate that airfoils with 6° deflection achieved the highest aerodynamic efficiency relative to the base model in the post-stall regime (e.g., α = 18°), with observed reductions in laminar separation bubbles (LSB) and trailing-edge vortex intensity. At moderate angles of attack (α = 10°), airfoils with 4° deflection exhibited optimal performance, yielding a 73.61% improvement in CL/CD compared to the base model. While stall characteristics revealed limited improvement, all modified geometries demonstrated reduced drag compared to the base model.Öğe Adana'da hava kirliliğinin halk üzerindeki etkilerifarkındalık ve bilinç düzeyinin değerlendirilmesi(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Turhan, Esra; Aydın, RozalinTüm dünyada, dolayısıyla Türkiye'de de çevresel faktörler nedeniyle hava kalitesinde meydana gelen hızlı bozulmalar, yalnızca insan yaşamını ve sağlığını olumsuz etkilemekle kalmayıp, aynı zamanda toprağa, ekinlere, ormanlara, göllere ve nehirlere de zarar vermektedir. Türkiye'de artan nüfus, yoğunlaşan kentleşme, ulaşım ve sanayileşme, hava kirliliği düzeylerinin giderek yükselmesine neden olmakta; sanayileşmenin artışıyla birlikte kırsal alanlardan kente göç de emisyonların artmasına katkıda bulunarak, hava kirliliğini hem ülke genelinde hem de özellikle Adana bölgesinde ciddi bir sağlık sorunu haline getirmiştir Bu doğrultuda, Adana ilinde hava kirliliğine ilişkin halkın farkındalık ve bilinç düzeylerini belirlemek amacıyla, 402 kişiye 35 soruluk yüz yüze anket uygulanmış, katılımcıların hava kirliliğinin algısı, kaynakları, etkileri ve buna yönelik bireysel tutumları değerlendirilmiştir. Elde edilen veriler SPSS programı kullanılarak analiz edilmiş; bulgular, katılımcıların büyük bir çoğunluğunun Adana'daki hava kirliliğinin farkında olduğunu, ancak kişisel önlem alma konusunda yetersiz davrandıklarını ortaya koymuştur. Anket sonuçlarına göre, katılımcıların %77'si hava kirliliğinin insan kaynaklı olduğuna inanırken, %68'i kirliliğin sağlık üzerinde olumsuz etkileri bulunduğunu belirtmiş; en sık bildirilen sağlık sorunları arasında nefes darlığı, göz rahatsızlıkları ve baş ağrısı yer almıştır. Mevsimsel analizlerde, hava kirliliğinin kış aylarında ve akşam saatlerinde daha yoğun hissedildiği, özellikle Seyhan ilçesinin en çok etkilenen bölge olduğu belirlenmiştir. Bireysel önlemler açısından, katılımcılar hava kirliliğinin kaynakları konusunda bilgi sahibi olmalarına rağmen, düzenli hava kalitesi ölçümleri için maddi katkıda bulunma veya korunma amacıyla maske kullanma konusunda isteksizlik göstermiş; ayrıca, ev seçimi sırasında hava kalitesinin göz ardı edilerek kira ve mülk fiyatının öncelikli kriter olarak belirlendiği gözlemlenmiştir. Ulaşım tercihlerine ilişkin bulgular ise, otobüs ve özel araç kullanımının yaygın olduğunu; otoyolların ulaşım kolaylığı sağlamakla birlikte, gürültü ve hava kirliliğini artırdığı şeklinde yorumlanmıştır. Bu çalışma, Adana'daki hava kirliliğine yönelik farkındalık ve bireysel tutumların belirlenmesi açısından önemli veriler sunmakta; elde edilen bulgular, çevresel duyarlılığın artırılması ve etkin politika stratejilerinin geliştirilmesi için temel oluşturacak niteliktedir.Öğe Business analytics with data mining: An investigation of web based data with sentiment analysis(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Özmen, Cemile Gökçe; Gündüz, SelimConsumers refer to product reviews in the processes of decision-making and obtaining information about the product before performing purchasing behaviour through e-commerce. Product reviews are produced by other consumers who have already purchased and experienced the product. It is aimed in this study to examine cosmetic products in a Turkey based e-commerce website with sentiment analysis and to create a new domain-specific Turkish sentiment dictionary model with manual labelling. In the study, a Turkish sentiment dictionary consisting of 65,378 words was created by manually labelling 875,445 product comments obtained from the web and sentiment analysis was performed using this dictionary. The data set is used for positive, neutral and negative classification problems by using various machine learning algorithms. Algorithms are compared by evaluated with accuracy, precision, recall and f-1 score metrics. The performance of the algorithms was highly successful in the groups and categories to which the product reviews were assigned. Compared to other algorithms, SVM showed the highest success in all categories. Thus, the created sentiment analysis dictionary showed classification success in the field of cosmetics and achieved high performance. The dictionary created in the study for the cosmetics sector is a reference source for similar or further studies to be carried out in the future.Öğe Yüksek biyoaktiviteye sahip kazein hidrolizatı üretim koşullarının optimizasyonu ve gıda katkı maddesi olarak kullanılabilirliği(Adana Alparslan Türkeş Bilim ve Teknoloji Üniversitesi, 2025) Çelik, Özlem; Erbay, ZaferBiyoaktif peptitler, proteinlerin hidrolizasyonu ile elde edilmektedirler. Süt proteinlerinden kazein, biyoaktif peptitler açısından öncül bir kaynak olarak bilinmektedir. Bu tez çalışmasında, yüksek biyoaktif peptit içeriğine sahip kazein proteinin maksimum aktivite değerlerinde optimum koşullarda üretilip gıda katkı maddesi olarak kullanılabilirliği gösterilmiştir. İlk olarak uygun endopeptidaz enzimine karar verilmiş (Alcalase®, Neutrase® ve rTrypsin®) sonrasında biyoaktivite etkinliğini arttırmak amacıyla ekzopeptidaz enzim kombinasyonları denenmiştir. En uygun endopeptidaz 12 saat inkübasyonda Neutrase ultrases ön işlem uygulanan hidrolizat koşulları olarak belirlenmiştir (toplam peptit miktarı 258,0 mg Tripton / g örnek, antihipertansif potansiyel %77,6, antioksidan potansiyel 686,8 µmol Troloks / mL örnek, metal bağlama %31,4, alfa amilaz inhibisyonu %67,1, lipaz inhibisyonu %40,9). Ekzopeptidaz kullanımında Protana™ Prime ve Flavourzyme® enzimlerinin eş zamanlı kullanılması uygun bulunmuştur (toplam peptit miktarı 447,2 mg Tripton / g örnek, antihipertansif potansiyel %50,5, antioksidan potansiyel 622,3 µmol Troloks / mL örnek, metal bağlama %34,1, alfa amilaz inhibisyonu %92,7, lipaz inhibisyonu %40,4). Optimizasyon faktörü olarak inkübasyon süresi, enzim konsantrasyonu ve ultrases süresi değerlendirilmiştir. En yüksek biyoaktivite değerleri 12 saat inkübasyon, 30 dakika ultrases ve 0,5 mL enzim konsantrasyonunda gözlemlenmiştir. Optimum koşullarda elde edilen yüksek biyoaktivite değerlerine sahip hidrolizat ayran örneklerinde %1, %5 ve %10 oranında eklenerek 1., 7. ve 14. günlerde bileşim, duyusal ve biyoaktivite analizleri değerlendirilip sonuç olarak %1 ve %5 konsantrasyonları tercih edilmiştir.
