Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/395
Browse
15 results
Search Results
Article Probing Adversarial Robustness of Protein Language Models: A Reproducible Case Study of ESM-2 Under Substitution-Based Attacks(AIUB Office of Research and Publication, 2026) Aydin, Zafer; Niloy, Md. Robiul Islam; Moazzam, Md. GolamArticle An Attention-Based Autoencoder Model with Gated Recurrent Unit for Stock Price Movement Prediction(Springer Nature, 2026) Kolukisa, Burak; Bakir-Gungor, Burcu; Akkas, HuseyinPredicting stock movements is crucial for investors looking to maximize their profits in rapidly changing financial markets. However, noise in stock price data makes it harder to detect the trends and ultimately decreases the performance of predictive models. To address these challenges that are caused by noise, this study proposes two novel models, i.e., an Attention-based Autoencoder (ABA) and an Attention-based Variational Autoencoder with Gated Recurrent Units (AGRUA), which uniquely integrate denoising, attention, and GRU layers. Firstly, a data set is created by adding 9 different technical indicators to the historical data of Borsa Istanbul (BIST 30) stocks. Secondly, proposed methods and other well-known deep learning models were used to remove noise from the data sets. Finally, each denoised dataset was fed separately to the Extreme Gradient Boosting model and subjected to a buy-sell process. The results were measured using trading indicators such as amount of the profits and Sharpe Ratio, Sortino Ratio and Maximum Drawdown. The proposed models produced substantial financial gains, with AGRUA achieving the highest total profit and ABA achieving the lowest average Maximum Drawdown, thereby demonstrating superior risk-adjusted performance. Lastly, Friedman and Nemenyi tests confirmed that AGRUA and ABA surpass most of the benchmarks in profit and risk-adjusted returns. The performance of the proposed method demonstrates its value in capturing nonlinear market patterns and improving decision accuracy, emphasizing the need for noise reduction before forecasting.Conference Object Citation - Scopus: 2Combining Classifiers for Protein Secondary Structure Prediction(Institute of Electrical and Electronics Engineers Inc., 2017-09) Aydin, Zafer; Uzut, Ömmu GülsümArticle Frequency-Based Deep Occlusion Awareness Instance Segmentation(MDPI, 2026-02-26) Guzel, Yasin; Aydin, Zafer; Talu, Muhammed FatihOne major challenge faced by deep learning-based methods that detect target objects in the form of bounding boxes is object occlusion. High degrees of occlusion significantly diminish the accuracy of instance segmentation. Nonetheless, complex-valued Fourier descriptors can robustly represent object boundaries using minimal information. In this study, the impact of integrating Fourier descriptors-renowned for their strong representational capacity-with deep network models (UNet) that exhibit high generalization performance on instance segmentation accuracy was investigated. Within the scope of the research, nine network models were designed based on different strategies for utilizing frequency components. These variants fall into four strategy families: (i) UNet-style spectrum regression on fixed low-frequency windows (FUNet), (ii) magnitude-guided frequency selection/ROI construction (FUNet-Thr, FUNet-BBox), (iii) sequence models over tokenized FFT coefficients (BiLSTM Patch/Sorted), and (iv) encoder-only spectrum predictors with different depth/capacity (EncoderFFT1/2). To fairly evaluate the models' performance in segmenting objects subjected to disruptive factors (e.g., occlusion, blurring, noise), a specialized synthetic dataset was prepared. The task is formulated as single-target (single-instance), single-class segmentation. This dataset, automatically generated according to initial parameter values, contains images of objects moving at various speeds within a single frame. Among these models, the one termed FUNet, which relies on partial matching of central frequency components, achieved the highest segmentation accuracy despite the disruptive effects. Under the challenging Dataset 8 setting, the proposed FUNet achieved the highest overlap-based performance (Dice = 0.9329, IoU = 0.8842) among Attention U-Net, U-Net, and FourierNet, with statistically significant gains confirmed by paired per-image tests.Article BrAIn: A Comprehensive Artificial Intelligence-Based Morphology Analysis System for Brain Organoids and Neuroscience(Wiley, 2026-03-12) Polatli, Elifsu; Guner, Huseyin; Bastanlar, Yalin; Karakulah, Gokhan; Evranos, Ali Eren; Kahveci, Burak; Guven, SinanHuman-induced pluripotent stem cells (iPSCs) offer transformative potential for biomedical research, with iPSC-derived organoids providing more physiologically relevant models than traditional 2D cell cultures. Among these, brain organoids (BO) are particularly valuable for drug screening, disease modeling, and investigations into molecular pathways. Accurate representation of brain morphology is critical, as more complex organoid structures better mimic the human brain. Deep learning (DL) and machine learning (ML) approaches have become integral to analyzing organoid morphology, yet tools for comprehensive, time-resolved assessments are scarce. Here, we introduce BrAIn, a DL-based application for analyzing the developmental progression of BOs. BrAIn tracks their evolution from embryoid bodies (EBs) and quantifies parameters including area, Feret diameter, perimeter, roundness, and circularity. It also classifies budding and abnormal morphologies of 3D organoids and detects monolayer neural rosette structures, key features of neuronal differentiation. Designed with accessibility in mind, BrAIn provides a no-code interface, enabling researchers of all technical backgrounds to conduct advanced morphological analyses with ease. Our study demonstrates the application of BrAIn to evaluate the effects of different growth conditions-static, orbital shaker, and microfluidic chip-based-on BO development. Orbital shaker cultures resulted in the largest organoids, while chip-based systems achieved more homogeneous growth. Both conditions produced organoids with greater morphological complexity compared to static culture. BrAIn emerges as a robust, user-friendly tool to quantify BO development and explore how versatile growth conditions influence their morphology and maturation.Article Citation - WoS: 1Citation - Scopus: 2Spec17Tre: A New Dataset in Hardware Security and Using Deep Learning for Detecting Spectre Attacks(Springer Heidelberg, 2025-05-21) Aktas-Aydin, Hatice; Yalcin, GulayComputer performance has become a significant subject of study due to the processing of big data, the complexity of calculations and the importance of time efficiency. Many companies are improving processor operating principles to increase performance. The most common methods for this purpose are speculative execution and cache usage. While these techniques improve performance, they also introduce certain security vulnerabilities. Spectre is an attack that exploits vulnerabilities created by speculative execution, affecting all modern processor architectures. Research has shown that using machine learning to detect these attacks can be quite effective, although the features are typically gathered at the software level, which may limit detection since some performance parameters are not conveyed to the software. This study presents an analysis of Spectre attacks and their detection using machine learning and deep learning methods at the hardware level. Experiments are conducted using GEM5, a full-system hardware simulator, to ensure that only hardware-visible performance parameters are also collected. Attack detection is performed using Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) methods. The LSTM method is used in conjunction with SVM and Convolutional Neural Network (CNN) techniques, and all models were tested on a new dataset, Spec17Tre, created using "519.lbm" from the SPEC CPU2017 benchmarks. The study achieved a 95% accuracy rate in attack detection using the LSTM + CNN hybrid model, which also yielded an F1 score of 0.999 for detecting applied Spectre attack scenarios.Conference Object Citation - Scopus: 4Re-Exploring the Kayseri Culture Route by Using Deep Learning for Cultural Heritage Image Classification Cultural Heritage Image Classification by Using Deep Learning: Kayseri Culture Route(Association for Computing Machinery, 2024-05-25) Kevseroğlu, Ozlem; Kurban, RifatThe categorization of images captured during the documentation of architectural structures is a crucial aspect of preserving cultural heritage in digital form. Dealing with a large volume of images makes this categorization process laborious and time-consuming, often leading to errors. Introducing automatic techniques to aid in sorting would streamline this process, enhancing the efficiency of digital documentation. Proper classification of these images facilitates improved organization and more effective searches using specific terms, thereby aiding in the analysis and interpretation of the heritage asset. This study primarily focuses on applying deep learning techniques, specifically SqueezeNet convolutional neural networks (CNNs), for classifying images of architectural heritage. The effectiveness of training these networks from scratch versus fine-tuning pre-existing models is examined. In this study, we concentrate on identifying significant elements within images of buildings with architectural heritage significance of Kayseri Culture Route. Since no suitable datasets for network training were found, a new dataset was created. Transfer learning enables the use of pre-trained convolutional neural networks to specific image classification tasks. In the experiments, 99.8% of classification accuracy have been achieved by using SqueezeNet, suggesting that the implementation of the technique can substantially enhance the digital documentation of architectural heritage. © 2024 Elsevier B.V., All rights reserved.Article Citation - WoS: 4Citation - Scopus: 10Probabilistic Assessment of Wind Power Plant Energy Potential Through a Copula-Deep Learning Approach in Decision Trees(Cell Press, 2024-04) Sahin, Kubra Nur; Sutcu, MuhammedIn the face of environmental degradation and diminished energy resources, there is an urgent need for clean, affordable, and sustainable energy solutions, which highlights the importance of wind energy. In the global transition to renewable energy sources, wind power has emerged as a key player that is in line with the Paris Agreement, the Net Zero Target by 2050, and the UN 2030 Goals, especially SDG-7. It is critical to consider the variable and intermittent nature of wind to efficiently harness wind energy and evaluate its potential. Nonetheless, since wind energy is inherently variable and intermittent, a comprehensive assessment of a prospective site's wind power generation potential is required. This analysis is crucial for stakeholders and policymakers to make well-informed decisions because it helps them assess financial risks and choose the best locations for wind power plant installations. In this study, we introduce a framework based on Copula-Deep Learning within the context of decision trees. The main objective is to enhance the assessment of the wind power potential of a site by exploiting the intricate and non-linear dependencies among meteorological variables through the fusion of copulas and deep learning techniques. An empirical study was carried out using wind power plant data from Turkey. This dataset includes hourly power output measurements as well as comprehensive meteorological data for 2021. The results show that acknowledging and addressing the non-independence of variables through innovative frameworks like the Copula-LSTM based decision tree approach can significantly improve the accuracy and reliability of wind power plant potential assessment and analysis in other real-world data scenarios. The implications of this research extend beyond wind energy to inform decision-making processes critical for a sustainable energy future.Article Citation - Scopus: 6Network Intrusion Detection Based on Machine Learning Strategies: Performance Comparisons on Imbalanced Wired, Wireless, and Software-Defined Networking (SDN) Network Traffics(Turkiye Klinikleri, 2024-07-26) Hacilar, Hilal; Aydin, Zafer; Güngör, Vehbi ÇağrıThe rapid growth of computer networks emphasizes the urgency of addressing security issues. Organizations rely on network intrusion detection systems (NIDSs) to protect sensitive data from unauthorized access and theft. These systems analyze network traffic to detect suspicious activities, such as attempted breaches or cyberattacks. However, existing studies lack a thorough assessment of class imbalances and classification performance for different types of network intrusions: wired, wireless, and software-defined networking (SDN). This research aims to fill this gap by examining these networks’ imbalances, feature selection, and binary classification to enhance intrusion detection system efficiency. Various techniques such as SMOTE, ROS, ADASYN, and SMOTETomek are used to handle imbalanced datasets. Additionally, eXtreme Gradient Boosting (XGBoost) identifies key features, and an autoencoder (AE) assists in feature extraction for the classification task. The study evaluates datasets such as AWID, UNSW, and InSDN, yielding the best results with different numbers of selected features. Bayesian optimization fine-tunes parameters, and diverse machine learning algorithms (SVM, kNN, XGBoost, random forest, ensemble classifiers, and autoencoders) are employed. The optimal results, considering F1-measure, overall accuracy, detection rate, and false alarm rate, have been achieved for the UNSW-NB15, preprocessed AWID, and InSDN datasets, with values of [0.9356, 0.9289, 0.9328, 0.07597], [0.997, 0.9995, 0.9999, 0.0171], and [0.9998, 0.9996, 0.9998, 0.0012], respectively. These findings demonstrate that combining Bayesian optimization with oversampling techniques significantly enhances classification performance across wired, wireless, and SDN networks when compared to previous research conducted on these datasets. © 2024 Elsevier B.V., All rights reserved.Article Citation - WoS: 28Citation - Scopus: 50Metabolic Imaging Based Sub-Classification of Lung Cancer(IEEE-Inst Electrical Electronics Engineers Inc, 2020) Bicakci, Mustafa; Ayyildiz, Oguzhan; Aydin, Zafer; Basturk, Alper; Karacavus, Seyhan; Yilmaz, BulentLung cancer is one of the deadliest cancer types whose 84% is non-small cell lung cancer (NSCLC). In this study, deep learning-based classification methods were investigated comprehensively to differentiate two subtypes of NSCLC, namely adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). The study used 1457 F-18-FDG PET images/slices with tumor from 94 patients (88 men), 38 of which were ADC and the rest were SqCC. Three experiments were carried out to examine the contribution of peritumoral areas in PET images on subtype classification of tumors. We assessed multilayer perceptron (MLP) and three convolutional neural network (CNN) models such as SqueezeNet, VGG16 and VGG19 using three kinds of images in these experiments: 1) Whole slices without cropping or segmentation, 2) cropped image portions (square subimages) that include the tumor and 3) segmented image portions corresponding to tumors using random walk method. Several optimizers and regularization methods were used to optimize each model for the diagnostic classification. The classification models were trained and evaluated by performing stratified 10-fold cross validation, and F-score and area-under-curve (AUC) metrics were used to quantify the performance. According to our results, it is possible to say that inclusion of peritumoral regions/tissues both contributes to the success of models and makes segmentation effort unnecessary. To the best of our knowledge, deep learning-based models have not been applied to the subtype classification of NSCLC in PET imaging, therefore, this study is a significant cornerstone providing thorough comparisons and evaluations of several deep learning models on metabolic imaging for lung cancer. Even simpler deep learning models are found promising in this domain, indicating that any improvement in deep learning models in machine learning community can be reflected well in this domain as well.