Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.14627/7
Browse
16 results
Search Results
Conference Object Low Error-Rate Temperature Prediction with Machine Learning Using LoRaWAN Data(Institute of Electrical and Electronics Engineers Inc., 2025) Nas, Abdurrahim; Yildiz, Omer; Karlik, Sait EserConference Object Classification of Noisy Histopathology Images with Bayesian and Ensemble Deep Learning(SPIE, 2025) Akturk, B.; Bilgin, G.The classification of histopathology image data using deep learning algorithms is a pivotal aspect in cancer diagnosis, offering crucial benefits in terms of accuracy and rapid decision-making for pathologists, technologists, clinicians, and researchers. However, the complex process of data structuring and labeling often involves the presence of noisy or outlier labels and images, which inevitably diminish the training and predictive efficacy of classification models. This study introduces an innovative approach for classifying histopathology images with noisy characteristics, presenting a learning process resilient to noise. In this dataset, where classification performance is inherently challenged, popular transfer learning models are integrated with Bayesian deep learning and ensemble learning algorithms. This integration helps mitigate common problems such as overfitting and underfitting in outlier data predictions, yielding more stable outcomes. These hybrid methods, unaffected by the noise and uncertainty in data structures, demonstrate performance comparable to, or even surpassing, transfer learning algorithms. Consequently, the viability of addressing real-world noisy dataset scenarios is enhanced. Experimental results indicate that the proposed ensemble Bayesian transfer learning algorithms match the success of pre-trained transfer learning algorithms and offer an alternative to time-consuming model validation techniques, underscoring the practical application of ensemble Bayesian transfer learning in enhancing uncertainty assessments. · © 2025 SPIE · 0277-786X ·Conference Object Psycho-Social Impact of the Disaster on Employees in Terms of Occupational Health and Safety: The Case of Turkey(Springer-Verlag Singapore Pte Ltd, 2025) Aytac, Sevinc Serpil; Akalp, Husre Gizem; Bilir, Burcu Ongen; Mamaci, MerveThe earthquakes that occurred in Maras, Turkey, on February 6, 2023, with intensities of 7.7 MW and 7.6 MW, deeply affected the lives of millions of people in economic, social and psychological terms within a few seconds and caused loss of life. The aim of this study is to examine the moderator role of disaster preparedness plans in workplaces on the relationship between post-traumatic stress levels and depression, anxiety, stress levels of people who are directly or indirectly exposed to these earthquakes and have an active work life. The data was obtained from a total of 206 blue-collar employees who were actively working in a private company and were directly and indirectly exposed to the effects of the earthquake. In data collection, demographic information form, The Impact of Event Scale-Revised (IES-R) and Depression Anxiety Stress Scale-21 (DASS-21) were used. According to the findings analyzed using structural equation modeling showed that having a disaster preparedness plan in workplaces weakens the strength of the relationship between post-traumatic stress level and depression, anxiety, stress levels.Conference Object Sts: AI-Driven Smart Test Scenario Generation Tool(IEEE, 2025) Baglum, Cem; Yayan, UgurOne of the most critical steps in the software testing lifecycle, test scenario generation, reduces process efficiency due to its high time and resource requirements. As an innovative solution to this issue, the Smart Test Scenario Tool (STS) has been developed. Smart Test Scenario Tool (STS) enhances contextual accuracy and automation in test scenario generation by analyzing documents in xlsx, py, cpp, txt, and docx formats using large language models. This approach minimizes time loss, and the risk of errors encountered in traditional manual testing processes while transforming test procedures into a context-driven and systematic framework, offering an innovative contribution to the literature. Strengthened with a Streamlit interface, MongoDB-supported database management, and Ollama integration, the system enables the test scenario generation process, a critical component of the software testing cycle, to be conducted more efficiently and reliably. The validity of the study was confirmed through two distinct projects, the first implemented in Python and the second in C++.Conference Object Reflection Coefficient Prediction in Triple-Layer Microwave Absorbers: A Machine Learning Perspective(Institute of Electrical and Electronics Engineers Inc., 2025) Nas, Abdurrahim; Kankilic, Sueda; Karpat, EsinElectromagnetic absorbers prevent the reflection and transmission of electromagnetic waves. Electromagnetic absorbers have a wide range of applications from military to medical applications. In these areas, absorber designs have different importance in terms of parameters such as reflection coefficient, selected material and thickness. Many difficulties are encountered to achieve the optimal design. In this paper, we propose a machine learning regression method for three-layer microwave absorber architecture to obtain the optimum parameters, overcome the difficulties and speed up the process. The material and thickness of each layer are used as parameters to feed the models and the reflection coefficient is estimated using these parameters. Predictions are made with various regression algorithms. These algorithms are KNeighbors Regression, Random Forest Regressor, XGBoost Regression, CatBoost Regressor, AdaBoost Regressor which uses similarities between observations, Gradient Boosting Regressor which is tree based or boosted tree based algorithms, Linear Regression which uses a linear model, Partial Least Squares Regression which uses cross decomposition, Gaussian Process Regressor which uses statistical distribution, Stochastic Gradient Descent Regressor which uses a linear model to reduce empirical loss to predict an output. Mean square error (MSE), mean absolute error (MAE), root mean square error (RMSE), R-squared (R2) are used with the predictions of each model to obtain the metrics for the analysis of the results. The predicted values and actual values of the metrics are used to compare the regression algorithms used in the research. After the comparison, our observations show that in most cases CatBoost Regressor is better than other models used in the research. In general, it is observed that most of the machine learning regression algorithms used in this paper can be used to predict the reflection coefficient of three-layer microwave absorbers as output and input parameters used in the research. © 2025 Elsevier B.V., All rights reserved.Conference Object A Unique Web 3 Product-NFT Usage in Healthcare(IEEE, 2025) Unozkan, HuseyinNon-fungible tokens (NFTs) have gained immense popularity and value in recent years. After the Web 3 revolution, many new products have been constructed based on smart contracts or powerful software agreements. One of the most exciting products of Web 3 is NFT, with their increasing popularity, high volatility, and significant price movements making them popular in trading activities. Although many people, investors and researchers perceive NFTs as a trading product, the smart contract and metadata, both of which make up of NFT structure, offer valuable support for the big data storage and processing fields. In recent years, researchers have proposed different uses of NFTs in healthcare systems to improve the quality of services by using NFTs' versatile data storage architecture. In this study, the capability of NFTs in the storage of health data is investigated, the usage and usage proposals of NFTs in health are examined, and probable usage areas in health sciences are evaluated. The uniqueness of this study is, using NFTs as the dead parental data storage in which the metadata does not need to be mitigated, but the stored data is very valuable for doctors to assess the genetic illnesses comes from parental DNAs.Conference Object Forty Years of Event Graphs in Research and Education(Institute of Electrical and Electronics Engineers Inc., 2023) Gunal, M.M.; Osais, Y.I.; Schruben, L.; Wagner, G.; Yücesan, E.Forty years ago, in 1983, Lee Schruben proposed the Event Graph formalism and modeling language, subsequently defining the paradigm of Event-Based Simulation, in a precise way, which had been pioneered 20 years before by SIMSCRIPT. The purpose of this panel is for a group of Event Graph researchers both from Operations Research and Computer Science, including the inventor of Event Graphs and one of his former PhD students who has made essential contributions to their theory, to discuss their views on the history and potential of Event Graph modeling and simulation. In particular, the adoption of Event Graphs as a discrete process modeling language in Discrete Event Simulation and in Computer Science, and their potential as a foundation for the entire field of Discrete Event Simulation and for the fields of process modeling and AI in Computer Science is debated. © 2023 IEEE.Conference Object Citation - Scopus: 2Event Graphs: Syntax, Semantics, and Implementation(Institute of Electrical and Electronics Engineers Inc., 2023) Gunal, M.M.; Ismail Osais, Y.; Wagner, G.This tutorial aims to introduce Event Graphs (EGs), invented 40 years ago by Lee Schruben to allow eventbased modeling of discrete dynamic systems. Their simplicity and naturalness in causality modelling and simulation modelling made EGs popular in research and practice. In a simulation, an event causes state changes in a system as well as other events to happen in the future. EGs provide a parsimonious diagram representation for the Event Scheduling paradigm of Discrete Event Simulation. We first introduce their visual syntax and informal semantics, and then present a recent extension by adding objects to EGs. Our tutorial also includes an introduction to the formal semantics of EGs and a Python implementation for executing EGs. © 2023 IEEE.Conference Object Transaction Types in Cryptocurrencies(Institute of Electrical and Electronics Engineers Inc., 2024) Uysal, T.; Unozkan, H.The cryptocurrency ecosystem has witnessed an explosive growth, and this study delves into the mechanisms that fuel this expansion. Focusing on airdrops, staking, farming, and coin burning, the research analyzes a vast dataset of transactions from the Ethereum and Binance Networks. This analysis sheds light on the strategic use of these tools by highlighting transactions that engage users and distribute rewards (e.g., airdrops, staking). Furthermore, the study investigates farming as a method to enhance market efficiency by providing liquidity, and coin burning as a strategy to manage token supply and potentially increase value through scarcity. While effective utilization of these mechanisms can bolster token value and project success, regulatory challenges remain. Ultimately, this study aims to raise public awareness of cryptocurrency transaction types and the associated risks. Although many researchers have studied illicit flows on cryptocurrency transactions, in literature we haven't confronted with any study regarding transaction types. By analyzing over 107 million transaction records, the research presents the distribution of these transaction types. With the analysis of the transaction types, the definitions of them, the statistical outputs from the more than 107 million transaction records from ERC20 and BEP20 blockchain systems and the analysis of some specific token types such as Pancake Swap and Shiba, this research is the first study to present an academic approach with statistical analysis. This study can provide investors with knowledge that safeguard them from the potential pitfalls of cryptocurrency transactions. Also, this study presents some basic statistics so as to understand main patterns of different types of transactions. © 2024 IEEE.Conference Object Citation - Scopus: 2Unet3D Based Next Frame Prediction;(Institute of Electrical and Electronics Engineers Inc., 2024) Akbacak, E.The concept of next-frame prediction, which is predicting the subsequent frames using historical frames' spatial and temporal properties, is indispensable in computer vision. There are various application of frame prediction such as predicting a future event in autonomous vehicles, predicting patient falls in biomedical engineering, and reducing the amount of data transmitted in video transmission. Deep learning applications in this field are the focus of the most effective methods. Especially CNN-LSTM, Convolutional LSTMs, and GAN-supported deep learning methods are very common. This study proposes the inflated 3D Unet encoder-decoder model, which is not yet used for the next-frame prediction problem. The proposed model predicts both the next frame and the subsequent frames. Experimental results have shown that the proposed method gives better results than CNN-LSTM and Convolutional LSTMs. © 2024 IEEE.