Sürdürülebilir Kalkınma Amaçları

• 2
• 2
• 2
• 1
• 1

• MakaleAçık Erişim2026Scopus SJRQ2/r/munzur/4034
  A New Metacognitive Confidence Calculation Approach Based on Large Language Models for Natural Language Explanations

  Yilmazer Gundem, Merve; Sahin, Simay; Karakose, Mehmet

  Traditional methods used to measure metacognitive confidence rely on superficial features such as word count and ambiguity and therefore fail to adequately reflect the level of metacognitive confidence in statements. In this study, in addition to traditional text-based confidence measures, we used a Large Language Model (LLM) to directly analyze confidence and its ratio in statements. LLM-based confidence scores were calculated on statements from the e-SNLI and CoS-E open-source datasets, and the prediction performance obtained with traditional features was compared. The results indicate that traditional surface-level text features explain the variance in LLM-generated confidence scores to a limited extent, and that LLM-based scores are significantly correlated with independent researcher evaluations. This study contributes to the literature by offering a scalable and reproducible analytical framework for measuring metacognitive confidence. It is believed that the findings can be directly applied, particularly in applications where trust is critical, such as automated software test scenario generation, decision support systems, and risk management. The use of LLMs offers a framework for the LLM-based analysis of potential metacognitive confidence signals in natural language explanations by examining the sense of confidence and certainty with which statements are presented, rather than their factual accuracy.

  İncele
• MakaleAçık Erişim2026Scopus SJRQ2/r/munzur/1051
  A Unified and Interpretable Benchmark of Classification Models for DGA-Based Power Transformer Fault Diagnosis

  Doğan, Kıvanç; Ertarğın, Merve; Çalışkan, Abuzer; Bakay, Ayşenur

  Power transformers are among the most critical components of electric power transmission and distribution systems, and unexpected failures can lead to substantial economic losses and prolonged power outages. Dissolved Gas Analysis (DGA) is the most widely used diagnostic technique for transformer fault diagnosis and involves interpreting gases dissolved in insulating oil. However, conventional interpretation methods, such as the Rogers ratio, the Doernenburg ratio, the IEC 60599 ratio method, and the Duval Triangle, rely heavily on expert knowledge, may produce inconsistent diagnoses for the same oil sample, and may fail to provide a diagnosis in certain cases. In this study, 12 classification models were evaluated using the publicly available Power Transformers Fault Detection and Diagnosis (FDD) and Remaining Useful Life (RUL) dataset and a unified evaluation protocol. Model performance was assessed using Accuracy, Balanced Accuracy, and Macro-F1 score, while model interpretability was investigated through Shapley Additive Explanations (SHAP) analysis. The results showed that ensemble tree-based methods achieved the best overall performance. LightGBM and Random Forest both attained an Accuracy of 0.969 and a Macro-F1 score of 0.924, while LightGBM further achieved the highest Balanced Accuracy of 0.940. XGBoost exhibited the most stable performance under cross-validation. SHAP analysis revealed that engineered relative concentration features, particularly the CO/H2 ratio and the combined gas ratio, were among the most influential features for fault classification. These findings demonstrate that, for datasets of this scale, ensemble tree-based models combined with well-designed features provide strong and interpretable performance for imbalanced DGA-based fault diagnosis, highlighting the effectiveness of feature-based ensemble learning for small- to medium-sized datasets.

  İncele
• MakaleAçık Erişim2026Scopus SJRQ2/r/munzur/811
  Hybrid Supervised Classification and Deep Embedding–Based Profiling Framework for Electricity Consumption Analysis

  Gunay, Mihriban; Yildirim, Ozal; Demir, Yakup; Zhilevski, Marin; Mikhov, Mikho; Yordanov, Nikolay

  This study proposes a hybrid deep learning framework that integrates supervised classification and unsupervised profiling for electricity consumption analysis. In the supervised phase, a one-dimensional Convolutional Neural Network combined with Long Short-Term Memory (1D CNN–LSTM) architecture is developed to classify daily load patterns. The performance of the proposed model is compared with traditional machine learning and deep learning approaches, including Support Vector Machine (SVM), k-Nearest Neighbors (KNN), a standalone Long Short-Term Memory (LSTM) model, a Transformer-based model, and a standalone 1D CNN model. Experimental results on the Precon house dataset and the CU-BEMS dataset demonstrate that the proposed hybrid architecture outperforms the benchmark models, achieving classification accuracies of 87.59% and 86.40%, respectively. In the unsupervised phase, the trained CNN–LSTM encoder is utilized as a deep feature extractor. The resulting 32-dimensional latent embeddings are clustered using K-Means, Gaussian Mixture Model (GMM), Agglomerative, Spectral, and Ensemble methods. Clustering robustness is evaluated through bootstrap-based stability analysis using the Adjusted Rand Index (ARI) and the Normalized Mutual Information (NMI). The results demonstrate stable and interpretable electricity consumption profiles, particularly in the residential dataset, where near-perfect clustering stability is observed for K-Means. The proposed framework provides both improved classification performance and robust consumption profiling based on deep embedding, offering a practical tool for energy management.

  • 
  • 
  • 
  AtıfScopus1·OpenAlex1

  İncele
• MakaleAçık Erişim2025Scopus SJRQ2/r/munzur/1316
  A New Ground-Motion Prediction Model for Shallow Crustal Earthquakes in Türkiye

  Çeken, Ulubey; Sertçelik, Fadime; İçen, Abdullah

  The recent expansion of the strong-motion observation network, along with the increase in data obtained during major earthquakes and efforts to create consistent metadata for source, path, and site effects for both old and new records, has significantly improved the quality of data and the level of modeling in Türkiye. The mainshock and aftershock records of the 6 February 2023 Kahramanmaraş earthquake doublet (MW 7.8 and 7.7), which are among the most destructive earthquakes in world history, constitute an up-to-date and important data source for this study. In this study, we present new ground-motion prediction models (GMPMs) for shallow crustal earthquakes using strong-motion data recorded in Türkiye. Our GMPMs are calibrated using 20,173 strong-motion records from 1565 shallow crustal earthquakes with depths of less than 35 km that occurred in Türkiye and its vicinity between 1976 and 2023. Our model is valid for magnitudes ranging from 4.0 to 7.8 (MW), and for the time-averaged 30 m shear wave velocity (VS30) values of 975 stations, which ranged from 131 to 1862 m/s. In the analyses performed, using the recently developed site amplification model, we calculated the model coefficients using the mixed-effects regression algorithms used by the GMPM developers. Additionally, a heteroscedastic model was created for aleatory variability as a function of MW. The closest distance to the surface projection of the fault plane (RJB) is between 0 and 350 km. Using the metadata prepared according to these criteria, we derived up-to-date ground-motion prediction models for horizontal-component peak ground velocity (PGV), peak ground acceleration (PGA), and 5% damped pseudo-spectral acceleration (PSA) response spectra, at 36 periods ranging from 0.01 to 10 s. The variability in the predictions was decomposed into within-event, between-event, and site-to-site deviations to determine the total standard deviations (σ). Compared to previous models, the proposed GMPMs were developed using a much richer database with recent major earthquakes, and the consistent estimates and lower residuals in the comparisons support the reliability of the models.

  AtıfScopus4·WoS3·OpenAlex4

  İncele
• MakaleAçık Erişim2025Scopus SJRQ2/r/munzur/1195
  Multiscale Time Series Modeling in Energy Demand Prediction: A CWT-Aided Hybrid Model

  Sezer, Elif; Yıldırım, Güngör; Özdemir, Mahmut Temel

  In the contemporary energy landscape, the increasing demand for electricity and the inherent uncertainties associated with the integration of renewable resources have rendered the accurate and reliable forecasting of short- and long-term demand imperative. Energy demand forecasting, fundamentally a time series problem, can be inherently complex, nonlinear, and multi-scale. Therefore, interest in artificial intelligence–based methods that provide high performance for short- and long-term forecasting, rather than traditional methods, has increased in order to solve these problems. In this study, a hybrid artificial intelligence model based on LSTM, GRU, and Random Forest, utilizing a distinct mechanism to address these types of problems, is proposed. The Multi-Scale Sliding Window (MSSW) approach was utilized for the model’s input data to capture the dynamics of the time series at different scales. The optimization of windows was conducted using the Continuous Wavelet Transform (CWT) method to determine the optimal window sizes within the MSSW structure in a data-driven manner. Experimental studies on Panama’s real energy demand data from 2015 to 2020 show that the CWT-aided MSSW-hybrid model forecasts better with lower error rates (0.007 MAE, 0.009 RMSE, 1.051% MAPE) than single models and manually determined window sizes. The results of the study demonstrate the importance of hybrid structures and window optimization in energy demand forecasting.

  • 
  • 
  AtıfScopus1·OpenAlex2

  İncele
• MakaleAçık Erişim2025Scopus SJRQ2/r/munzur/873
  A New, Robust, Adaptive, Versatile, and Scalable Abandoned Object Detection Approach Based on DeepSORT Dynamic Prompts, and Customized LLM for Smart Video Surveillance

  Yilmazer, Merve; Karakose, Mehmet

  Video cameras are one of the important elements in ensuring security in public areas. Videos inspected by expert personnel using traditional methods may have a high error rate and take a long time to complete. In this study, a new deep learning-based method is proposed for the detection of abandoned objects, such as bags, suitcases, and suitcases left unsupervised in public areas. Transfer learning-based keyframe detection was first performed to remove unnecessary and repetitive frames from the ABODA dataset. Then, human and object classes were detected using the weights of the YOLOv8l model, which has a fast and effective object detection feature. Abandoned object detection is achieved by tracking classes in consecutive frames with the DeepSORT algorithm and measuring the distance between them. In addition, the location information of the human and object classes in the frames was analyzed by a large language model supported by prompt engineering. Thus, an explanation output regarding the location, size, and estimation rate of the object and human classes was created for the authorities. It is observed that the proposed model produces promising results comparable to the state-of-the-art methods for suspicious object detection from videos with success metrics of 97.9% precision, 97.0% recall, and 97.4% f1-score.

  AtıfScopus5·WoS2·OpenAlex8

  İncele
• MakaleAçık Erişim2025Scopus SJRQ2/r/munzur/833
  A New Hybrid ConvViT Model for Dangerous Farm Insect Detection

  Utku, Anil; Kaya, Mahmut; Canbay, Yavuz

  This study proposes a novel hybrid convolution and vision transformer model (ConvViT) designed to detect harmful insect species that adversely affect agricultural production and play a critical role in global food security. By utilizing a dataset comprising images of 15 distinct insect species, the suggested approach combines the strengths of traditional convolutional neural networks (CNNs) with vision transformer (ViT) architectures. This integration aims to capture local-level morphological features effectively while analyzing global spatial relationships more comprehensively. While the CNN structure excels at discerning fine morphological details of insects, the ViT’s self-attention mechanism enables a holistic evaluation of their overall configurations. Several data preprocessing steps were implemented to enhance the model’s performance, including data augmentation techniques and strategies to ensure class balance. In addition, hyperparameter optimization contributed to more stable and robust model training. Experimental results indicate that the ConvViT model outperforms commonly used benchmark architectures such as EfficientNetB0, DenseNet201, ResNet-50, VGG-16, and standalone ViT, achieving a classification accuracy of 93.61%. This hybrid approach improves accuracy and strengthens generalization capabilities, delivering steady performance during training and testing phases, thereby increasing its reliability for field applications. The findings highlight that the ConvViT model achieves high efficiency in pest detection by integrating local and global feature learning. Consequently, this scalable artificial intelligence solution can support sustainable agricultural practices by enabling the early and accurate identification of pests and reducing the need for intensive pesticide use.

  • 
  AtıfScopus9·WoS5·OpenAlex11

  İncele
• MakaleAçık Erişim2024Scopus SJRQ2/r/munzur/2085
  The Behavior of Ceramic Fiber Geopolymer Concrete under the Effect of High Temperature

  Dalğıç, Aras; Yılmazer Polat, Berivan

  Geopolymer concrete (GC), also known as green concrete, contains slag, silica fume, and fly ash as binders. The absence of cement in concrete is critical to protect the world from the environmental impacts of cement production. In addition, exposure to high temperatures is a critical parameter that causes loss of strength in concrete. In this study, Geopolymer concrete samples were prepared with 10 different samples containing different proportions of slag, silica fume, and porous ash and subjected to various physical, mechanical, and optical tests. The sample (GS90) with optimum workability and compressive strength, which also showed high performance in water absorption, freeze-thaw, and UPV tests, was used in high-temperature tests. Portland cement concrete (PCC) was used as a control sample. This study investigated the effect of high temperatures on the physical and mechanical properties of fiber-free GCs containing 2%, 5%, and 10% by volume of ceramic fibers. Therefore, fiber-reinforced, fiber-free, and PCC specimens were subjected to high-temperature tests at 100, 300, 600, and 900 °C. As a result of the observation of crack growth, color changes, and compressive strength parameters in the samples subjected to high-temperature tests, the thermal resistance of the 10% ceramic fiber geopolymer concrete sample was 2.5% higher than other samples. There is no study in the literature that examines the behavior of ceramic fiber-reinforced geopolymer concrete at high temperatures. This research revealed an important finding by proving that ceramic fiber reinforcement increases the compressive strength of geopolymer concretes at a remarkable rate after high-temperature impact.

  • 
  AtıfScopus8·WoS10·OpenAlex8

  İncele
• MakaleAçık Erişim2024Scopus SJRQ2/r/munzur/1793
  Detection of Threats to Farm Animals Using Deep Learning Models: A Comparative Study

  Korkmaz, Adem; Agdas, Mehmet Tevfik; Kosunalp, Selahattin; Iliev, Teodor; Stoyanov, Ivaylo

  The increasing global population and environmental changes pose significant challenges to food security and sustainable agricultural practices. To overcome these challenges, protecting farm animals and effectively detecting potential environmental threats is critical for economic and ecological sustainability. In this context, the current study examined the animal detection capabilities and efficiency of advanced deep learning models, such as YOLOv8, Yolo-NAS, and Fast-RNN, across a dataset of 2462 images encompassing various animal species that could pose a risk to farm animals. After converting the images into a standardized format, they were divided into three sets for training, validation, and testing, and each model was evaluated on this dataset during the analysis process. The findings indicated that the YOLOv8 model demonstrated superior performance, with 93% precision, 85.2% recall, and 93.1% mAP50 values, while Yolo-NAS was particularly noteworthy for its high recall value, indicating a remarkable detection ability. The Fast-RNN model also offered significant efficiency with balanced performance. The results reveal the considerable potential of deep learning-based object detection technologies in protecting farm animals and enhancing farm security. Additionally, this study provides valuable insights for future model optimization and customization research.

  • 
  AtıfScopus9·WoS7·OpenAlex7

  İncele
• MakaleAçık Erişim2022Scopus SJRQ2/r/munzur/3022
  Temperature-Dependent Density and Magnetohydrodynamic Effects on Mixed Convective Heat Transfer along Magnetized Heated Plate in Thermally Stratified Medium Using Keller Box Simulation

  Ullah, Zia; Akkurt, Nevzat; Alrihieli, Haifaa F.; Eldin, Sayed M.; Alqahtani, Aisha M.; Hussanan, Abid; Ashraf, Muhammad; Jabeen, Mah

  The heat transmission properties along the non-magnetized geometries have been numerically obtainedby various researchers. These mechanisms are less interesting in engineering and industrial processes because of excessive heating. According to current studies, the surface is magnetized and the fluid is electrically conductive, which helps to lessen excessive surface heating. The main objective of the current analysis is to numerically compute the temperature-dependent density effect on magnetohydrodynamic convective heat-transfer phenomena of electrical-conductive fluid flow along the vertical magnetized and heated plate placed in a thermally stratified medium. For the purpose of numerical analysis, the theoretical process governing heat and magnetic intensity along a vertical magnetic plate is examined. By using suitable and well-known similarity transformations for integration, the non-linear coupled PDEs for the aforementioned electrical-conductive fluid flow mechanism are changed and subsequently converted into non-similar formulation. The Keller Box method is used to numerically integrate the final non-similar equations. The MATLAB software program plots the transformed algebraic equations graphically and quantitatively. The behavior of the physical quantities such asvelocity graph, magnetic field graph, and temperature plot along with their slopes that arerate of skin friction, the rate of heat transfer, and the rate of magnetic intensity for different parameters included in the flow model. The novelty of the current work is to compute the magneto-thermo analysis of electrically conducting flow along the vertical symmetric heated plate. First, we secure the numerical solution for steady part and then these results are used to find skin friction, heat transfer, and magnetic intensity. In the current work, the fluid becomes electrically conducing due to a magnetized surface which insulates heat during the mechanism and reduces the excessive heating. The results are excellent and accurate because they are satisfied by its given boundary conditions. Additionally, the current problems have a big impact on the production of polymer materials, glass fiber, petroleum, plastic films, polymer sheets, heat exchangers, catalytic reactors, and electronic devices.

  AtıfScopus20·OpenAlex18

  İncele