Osmaniye Korkut Ata University Academic Repository
Not a member yet
5726 research outputs found
Sort by
Explainable AI Framework for Software Defect Prediction
No full textSoftware engineering plays a critical role in improving the quality of software systems, because identifying and correcting defects is one of the most expensive tasks in software development life cycle. For instance, determining whether a software product still has defects before distributing it is crucial. The customer's confidence in the software product will decline if the defects are discovered after it has been deployed. Machine learning-based techniques for predicting software defects have lately started to yield encouraging results. The software defect prediction system's prediction results are raised by machine learning models. More accurate models tend to be more complicated, which makes them harder to interpret. As the rationale behind machine learning models' decisions are obscure, it is challenging to employ them in actual production. In this study, we employ five different machine learning models which are random forest (RF), gradient boosting (GB), naive Bayes (NB), multilayer perceptron (MLP), and neural network (NN) to predict software defects and also provide an explainable artificial intelligence (XAI) framework to both locally and globally increase openness throughout the machine learning pipeline. While global explanations identify general trends and feature importance, local explanations provide insights into individual instances, and their combination allows for a holistic understanding of the model. This is accomplished through the utilization of Explainable AI algorithms, which aim to reduce the black-boxiness of ML models by explaining the reasoning behind a prediction. The explanations provide quantifiable information about the characteristics that affect defect prediction. These justifications are produced using six XAI methods, namely, SHAP, anchor, ELI5, LIME, partial dependence plot (PDP), and ProtoDash. We use the KC2 dataset to apply these methods to the software defect prediction (SDP) system, and provide and discuss the results
A study on determination of chlorophyll-a values using remote sensing method
Get PDFBu çalışma kapsamında Osmaniye ilinin farklı lokalitelerinden mısır ve ayçiçeği yaprak örnekleri toplanarak spektrofotometrik olarak ve uzaktan algılama yöntemi ile klorofil içeriklerinin belirlenip belirlenemeyeceği araştırılmıştır. Araştırma sonunda, spektrofotometrik analizden elde edilen sonuçlara göre, Osmaniye ilinin Nohuttepe mevkiinden toplanan mısır bitkisinin hem klorofil-a (3.63 mg/g ya) hem de klorofil-b (1.23 mg/g ya) miktarının en yüksek düzeyde olduğu, Karataş, Cevdetiye mevkiinden toplanan ayçiçeği bitkisinin ise hem klorofil-a (0.3 mg/g ya) hem de klorofil-b (0.01 mg/g ya) miktarının en düşük seviyede olduğu tespit edilmiştir. Uzaktan algılama yöntemi kullanılarak klorofil-a miktarları Normalize Edilmiş Klorofil İndeksi (Normalized Difference Chlorophyll Index (NDCI)) ile belirlenmiştir. Ancak NDCI ile elde edilen klorofil-a miktarları 1,476-1,479 değerleri arasında değiştiği görülmüştür. Bu çalışmanın sonucunda uzaktan algılama yöntemi ile nokta bazında klorofil-a değerlerinin elde edilemediği sonucuna varılmıştır.In this study, corn and sunflower leaf samples were collected from different localities of Osmaniye province and their chlorophyll contents were investigated spectrophotometrically and by remote sensing method. At the end of the research, according to the results obtained from spectrophotometric analysis, it was determined that the corn plant collected from Nohuttepe location of Osmaniye province had the highest levels of both chlorophyll-a (3.63 mg/g fw) and chlorophyll-b (1.23 mg/g fw), while the sunflower plant collected from Cevdetiye location of Karataş had the lowest levels of both chlorophyll-a (0.3 mg/g fw) and chlorophyll-b (0.01 mg/g fw). Chlorophyll-a amounts were determined using the remote sensing method with the Normalized Difference Chlorophyll Index (NDCI). However, it was observed that the chlorophyll-a values obtained with the NDCI varied between 1.476 and 1.479. As a result of this study, it was concluded that point-based chlorophyll-a values could not be obtained by the remote sensing method
The impact of deep learning on diagnostic performance in the differentiation of benign and malignant thyroid nodules
No full textAims: This study aims to use deep learning (DL) to classify thyroid nodules as benign and malignant with ultrasonography (US). In addition, this study investigates the impact of DL on the diagnostic success of radiologists with different experiences. Material and methods: This study included 576 US images of thyroid nodules. The dataset was divided into 80% training and 20% test sets. Four radiologists with different levels of experience classified the images in the test set as benign-malignant. A DL model was then trained with the train set and predicted benign-malignant for the test set. Then, the output of the DL model for each nodule in the test set was presented to 4 radiologists, who were asked to make a benign-malignant classification again considering these DL results. Results: The accuracy of the DL model was 0.9391. The accuracy for junior resident (JR) 1, JR 2, senior resident (SR), and senior radiologist (Srad) before DL-assisting were 0.7043, 0.7826, 0.8435, and 0.8522 respectively. The accuracy in DL-assisted classifications was 0.9130, 0.8696, 0.9304, and 0.9043 for JR 1, JR2, SR, and Srad, respectively. DL assistance changed the decisions of less experienced radiologists more than more experienced radiologists. Conclusion: The DL model has superior accuracy in classifying thyroid nodules as benign-malignant with US images than radiologists with different levels of experience. Additionally, all radiologists, and most notably less experienced radiology residents, increased their accuracy in DL-assisted predictions
Will or be going to? Formulaic Patterns of Future Marking in L2 Academic English
No full textThis study examines how learners from different first language backgrounds express future time in academic English writing, focusing on will and be going to constructions. Drawing on a corpus of texts from 25 language backgrounds, the research reveals a strong preference for will (96.2%) over be going to (3.8%) across all groups. While will appears uniformly throughout texts, be going to is used more selectively, indicating distinct functional roles. The study documents variations in future expression density per text (2.30-4.85) and systematic differences in how future markers combine with different parts of speech, with Chinese learners showing particularly distinctive patterns. Cluster analysis reveals language family groupings, suggesting both universal constraints and L1 influences shape future expression. This finding shows will functions as a formulaic pattern in academic writing, advancing knowledge of how L2 learners develop temporal expressions
Driving the Built Environment Twin Transition: Synergising Circular Economy and Digital Tools
No full textThis chapter offers a comprehensive analysis of the intersection between digitalisation and the circular economy (CE) within the construction sector. It underscores the transformative potential of integrating digital tools to advance circularity objectives across managerial, environmental, economic, and social dimensions. The chapter discusses fourteen digital tools and technologies, which play a pivotal role in CE by streamlining data integration and visualisation, enhancing the accuracy of Life Cycle Costing (LCC) and Life Cycle Analysis (LCA) assessments, and supporting the adoption of CE strategies. Moreover, it explores how digital tools can facilitate collaboration among stakeholders, fostering knowledge sharing and effective communication throughout the project lifecycle. Nevertheless, challenges such as the absence of standardised methods, data interoperability issues, and the need for well-defined system boundaries remain. The chapter highlights the critical role of digitalisation in advancing the transition towards CE in the construction sector, emphasising the necessity of overcoming technical and systemic obstacles to fully harness the potential of digital tools in implementing CE. This transition aligns with the broader ambitions of the European Green Deal and the EU Digital Strategy, aiming to create a more sustainable, efficient, and resilient construction industry. By addressing these challenges and leveraging digitalisation, the construction sector can make a significant contribution to a sustainable and circular economy, ultimately benefiting both the environment and society. © The Author(s) 2025.European Commission, E
Microplastics in soil: a comprehensive review of analytical techniques
No full textMicroplastics (MPs) pollution has increasingly been recognized as a critical environmental issue impacting terrestrial ecosystems, particularly soil matrices. This review comprehensively evaluates existing identification techniques for MPs in soil, highlighting the complexities associated with soil matrices, such as heterogeneity, organic matter content, and diverse particle sizes. Current methods, including sieving, filtration, density separation, chemical digestion, and spectroscopic analysis (e.g., FTIR, Raman spectroscopy), are critically assessed for efficiency, reliability, and applicability. Our analysis identifies significant methodological inconsistencies across studies, emphasizing the urgent need for standardized analytical protocols to enable reliable comparative assessments. Recommendations include the implementation of stringent quality assurance/quality control measures to mitigate cross-contamination and enhance data quality. Given the projected increase in global plastic production and consequent MPs pollution, it is imperative to develop standardized, scalable, and cost-effective methodologies for monitoring MPs in soil environments.European Cooperation in Science and Technology10.13039/501100000921; COST (European Cooperation in Science and Technology)This article/publication is based upon work from COST Action CA20101 Plastics monitoRIng detectiOn RemedIaTion recoverY-PRIORITY, supported by COST (European Cooperation in Science and Technology), www.cost.eu
Akarsu Ortalama Akımlarının Çeşitli Makine Öğrenme Algoritmaları Kullanılarak Tahmini: Köprüçay Örneği
No full textAkarsu ortalama akımları havzanın su kaynaklarının yeterliliği hakkında önemli ipuçları barındırmaktadır. İklim değişikliği ile birlikte yağış ve sıcaklık gibi akarsu akımlarını doğrudan ilgilendiren parametrelerde bölgesel değişimler yaşanmaktadır. Yaşanan bu değişimler ortalama akımlarda da bölgesel farklılıklar görülmesine neden olmaktadır. Bu çalışmada Elektrik İdaresinin kayıtlarını paylaştığı Antalya ili Serik İlçesi Beşkonak Bucağında yer alan Köprüçay istasyonuna ait ortalama akımlar incelenmiştir. İstasyona ait 1957-2011 yılları arasındaki ortalama akımlar Multi-Layer Perceptron (MLP), Destek Vektör Makinaları (DVM) ve Random Forest (RF) makine öğrenme algoritmaları ile modellenmiştir. Çalışma iki kısımdan oluşmaktadır. İlk kısımda 1957-2011 yılları arasındaki veriler hem eğitim hem test kümesi olarak kullanılmış en uygun algoritmaya bu şekilde karar verilmiştir. İkinci kısımda algoritma seçiminden sonra kayıtları mevcut olmayan 2012-2022 yılları arasındaki ortalama akımlar tahmin edilmiştir. Modellemelerde ülkemize ait yıllık ortalama maksimum, minimum, ortalama sıcaklık ve ortalama yağış verileri girdi olarak kullanılmıştır. Sonuç olarak Köprüçay özelinde ortalama akım tahmininde en uygun algoritmanın RF olacağı görülmüştür
EFFECT OF REINFORCEMENT CONTENT AND CUTTING PARAMETERS ON TOOL WEAR IN MACHINING OF ALUMINUM HYBRID COMPOSITES
No full textThe machinability of aluminum hybrid composites (AHCs) can be enhanced by utilization of optimum composition and cutting parameters. In this study, the machinability ofAHCs containing micron-sized TiB2 and B4C particles was investigated using the face milling operation with a double coated cemented carbide tool. The composites, fabricated via cold pressing and microwave sintering, were subjected to the face milling experiments using a CNC milling machine. The influence of hybrid reinforcement content, feed rate and cutting speed on tool wear and surface roughness was examined during the milling of these composites. After the machinability tests, the worn surface of inserts was examined by a scanning electron microscope to see the wear types. The reinforcement content and cutting speed were obtained to have a much greater effect on the machinability and surface roughness of the hybrid composites compared to the feed rate. Either increasing the feed rate or decreasing the cutting speed provided a larger amount of chip removal until the wear limit. Moreover, the feed rate was obtained to be more effective on the tool wear at lower and amounts of reinforcement
The predictive role of attachment styles and decision-making styles in romantic relationship satisfaction
Get PDFBu çalışma, üniversite öğrencilerinin romantik ilişki doyumunu etkileyen başlıca psikolojik faktörlerden bağlanma stilleri ve karar verme tarzlarının yordayıcı rolünü incelemeyi amaçlamaktadır. Sosyal ilişkilerin bireylerin psikolojik iyi oluşu üzerindeki etkisi bilinse de romantik ilişkiler bu çerçevede ayrı bir önem taşımaktadır. Bu araştırmada güvenli, kaygılı ve kaçıngan bağlanma stilleri ile rasyonel, sezgisel, kaçınmacı ve panikleyici karar verme tarzlarının ilişki doyumuna olan etkileri birlikte ele alınmıştır. Araştırma ilişkisel tarama modeliyle yürütülmüş; veriler Kişisel Bilgi Formu, İlişki Doyumu Ölçeği, Üç Boyutlu Bağlanma Stilleri Ölçeği ve Melbourne Karar Verme Stilleri Ölçeği-II kullanılarak 485 üniversite öğrencisinden toplanmıştır. Analizlerde Pearson Korelasyon, Standart ve Hiyerarşik Çoklu Regresyon teknikleri uygulanmış; normallik, doğrusallık ve çoklu doğrusal bağlantı gibi varsayımlar test edilmiştir. Bulgular, bağlanma ve karar verme stillerinin birlikte romantik ilişki doyumundaki varyansın %53'ünü anlamlı şekilde açıkladığını göstermiştir. Güvenli bağlanma ve rasyonel karar verme doyumu pozitif, kaygılı ve kaçıngan bağlanma ile panikleyici ve kaçınan karar verme tarzları ise negatif yönde yordayıcı olarak bulunmuştur. Hiyerarşik regresyon sonucunda, bağlanma stillerinin tek başına %28.6 oranında varyans açıkladığı, karar verme tarzlarının eklenmesiyle bu oranın %53'e ulaştığı görülmüştür. Sonuçlar, bağlanma biçimleri ve karar stratejilerinin romantik ilişki doyumunu anlamlı biçimde etkilediğini ortaya koymaktadır.This study explores the predictive roles of attachment styles and decision-making styles on university students' romantic relationship satisfaction. While social relationships are known to impact psychological well-being, romantic relationships hold particular importance. This study investigates how secure, anxious, and avoidant attachment styles, along with rational, intuitive, avoidant, and panicked decision-making styles, jointly affect relationship satisfaction. Using a relational survey model, data were collected from 485 university students through the Personal Information Form, Relationship Satisfaction Scale, Three-Dimensional Attachment Styles Scale, and Melbourne Decision-Making Styles Scale-II. Analyses included Pearson Correlation, Standard Multiple Regression, and Hierarchical Multiple Regression, with assumptions such as normality, linearity, and multicollinearity tested. The findings indicated that attachment and decision-making styles together significantly explained 53% of the variance in relationship satisfaction. Secure attachment and rational decision-making were positive predictors, while anxious/avoidant attachment and panicked/avoidant decision-making were negative predictors. Hierarchical regression showed that attachment styles alone explained 28.6% of the variance, increasing to 53% with decision-making styles included. These results highlight the significant influence of attachment and decision-making patterns on romantic relationship satisfaction
Ultraviolet radiation application to fish fillet
Get PDFBu çalışmada, UV-C ışınlamanın çipura (Sparus aurata) balığı filetolarının mikrobiyal yükü, renk, pH, TVB-N içeriği ve tekstürel özellikleri üzerindeki etkileri araştırılmıştır. Mikrobiyolojik analizler, UV-C uygulaması, aerobik koloni sayısını azaltmış ve depolama süresince mikrobiyal gelişimi yavaşlatmıştır. Koliform, Enterobacteriaceae, Escherichia coli ve Vibrio cholerae bakterilerinin doğal yükleri başlangıçta düşük (<10 kob/g) ve hem kontrol hem de UV-C uygulanan grupta depolama süresince tespit edilmemiştir. Aerobik koloni sayısı kontrol örneklerinde 5,20×10³'ten 1,20×10? kob/g'a yükselirken, UV işlem görmüş örneklerde artış daha düşük kalmıştır (1,30×10?'ten 1,10×10? kob/g'a). Renk analizleri, UV işleminin başlangıçta L*, a* ve b* değerlerini etkilemediğini (P<0,05), ancak depolama süresince renk stabilitesini koruduğunu göstermiştir. pH değerleri arasında anlamlı fark bulunmamıştır. TVB-N sonuçları, kontrol grubunda 9. gün sonunda 30,17 mg/100g'a yükselirken, UV uygulanan örneklerde 17,61 mg/100g olarak ölçülmüş; bu da UV-C ışınlamanın bozulmaya bağlı azotlu bileşik oluşumunu önemli ölçüde engellediğini ve anlamlı düzeyde düşüş gözlemlenmiştir. (P<0,05). Tekstürel analizlerde UV-C işlemi, filetoların esnekliğini artırmış (7,36 mm - 9,73 mm) ve bu durum ürünün tazelik ile kalite düzeyini yükseltmiştir. Ayrıca Sertlik değerlerindeki koruma (13,59 N - 15,69 N), filetoların yapısal bütünlüğünün ve dokusal özelliklerinin korunduğunu gösterir. Diğer tekstürel parametrelerde anlamlı bir değişiklik gözlemlenmemiştir. Sonuç olarak, UV-C ışınlama, filetoların mikrobiyal güvenliğini ve kalite özelliklerini artırarak raf ömrünü uzatmada etkili bir yöntemdir.This study investigated the effects of UV-C irradiation on the microbial load, color, pH, TVB-N content, and textural properties of gilthead seabream (Sparus aurata) fillets. Microbiological analyses revealed that UV-C treatment significantly reduced aerobic colony counts and slowed microbial growth during storage. The initial natural loads of coliforms, Enterobacteriaceae, Escherichia coli, and Vibrio cholerae were low (<10 CFU/g), and none were detected throughout storage in either the control or UV-C-treated groups. Aerobic colony counts increased from 5.20×10³ to 1.20×10? CFU/g in control samples, whereas UV-C-treated fillets exhibited a less pronounced increase (from 1.30×10? to 1.10×10? CFU/g). Color analyses showed that UV-C did not affect the L*, a*, and b* values initially (P < 0.05), but helped maintain color stability over storage. No significant differences were observed in pH values between groups. TVB-N values in the control group reached 30.17 mg/100 g by day 9, while UV-C-treated fillets measured 17.61 mg/100 g, indicating that UV-C irradiation significantly inhibited the formation of nitrogenous compounds associated with spoilage (P < 0.05). Textural analysis demonstrated that UV-C treatment enhanced fillet elasticity (increasing from 7.36 mm to 9.73 mm), suggesting improved freshness and quality. The maintenance of hardness (13.59 N to 15.69 N) indicated preserved structural integrity and texture. No significant changes were observed in other textural parameters. Overall, UV-C irradiation proved to be an effective method for extending the shelf life of fish fillets by enhancing microbial safety and quality attributes