-
Fakülteler44
-
Bilgisayar Mühendisliği Bölümü44
-
Mühendislik Fakültesi44
-
Bilgisayar Yazılımı Anabilim Dalı23
-
Bilgisayar Mühendisliği22
-
Bilgisayar Donanımı Anabilim Dalı17
-
Bilgisayar Bilimleri Anabilim Dalı12
-
Elektrik - Elektronik Mühendisliği Bölümü3
-
Bilgisayar Bilimleri Anabilim Dalı3
-
Elektrik - Elektronik Mühendisliği3
Time and Memory Comparison of Parallel K-Nearest Neighbor Algorithms on GPUs
Korucu, Gizen Mutlu | Acı, Çiğdem
2022-11-18
An Adaptive Underwater Image Enhancement Framework Combining Structural Detail Enhancement and Unsupervised Deep Fusion
Kahveci, Semih | Avaroğlu, Erdinç
The underwater environment severely degrades image quality by absorbing and scattering light. This causes significant challenges, including non-uniform illumination, low contrast, color distortion, and blurring. These degradations compromise the performance of critical underwater applications, including water quality monitoring, object detection, and identification. To address these issues, this study proposes a detail-oriented hybrid framework for underwater image enhancement that synergizes the strengths of traditional image processing with the powerful feature extraction capabilities of unsupervised deep learning. Our framework introduces a novel multi-scale detail enhancement unit to accentuate structural information, followed by a Latent Low-Rank Representation (LatLRR)-based simplifi...
Optimization of Acoustic Entropy Source for Random Sequence Generation Using an Improved Grey Wolf Algorithm
Avaroğlu, Erdinç | Kahveci, Semih | Akkurt, Ramazan
The functionality of cryptographic systems necessitates unpredictable, high-quality random numbers. High-quality random numbers must possess unpredictability, non-reproducibility, and strong statistical properties. To achieve these qualities, True Random Number Generators (TRNG) are employed. The randomness quality of TRNG-derived sources depends on the entropy source used. Physical noise sources, ring oscillators, metastable, acoustic sources, and chaotic attractors are commonly used as entropy sources. In recent years, the use of acoustic signals as entropy sources has attracted attention. However, the noise in the signals affects the bit sequence to be generated. In addition, the threshold and sampling interval applied to the frequency values obtained from the signals also determine the...
A unified workflow strategy for analysing large-scale TripAdvisor reviews with BOW model
Bektaş, Jale
Nowadays, firms need to transform customer online reviews data properly into information to achieve goals such as having a competitive edge and improving the quality of service. This paper presents a unified workflow to solve the problems of analysing large-scale data with 710,450 reviews for 1,134 hotels by using text mining methods among the different touristic regions of Turkey. Firstly, a star schema dimensional data mart is built that includes one fact table and two dimensional tables. Then, a series of text mining processes which includes data cleaning, tokenisation, and analysis are applied. Text mining is implemented through standard BOW and the extended BON model. The results show significant findings through this workflow. We propose to build a dimensional model dataset before pe...
SVM-SMO-SGD: A hybrid-parallel support vector machine algorithm using sequential minimal optimization with stochastic gradient descent
Mutlu, Gizen | Acı, Çiğdem
The Support Vector Machine (SVM) method is one of the popular machine learning algorithms as it gives high accuracy. However, like most machine learning algorithms, the resource consumption of the SVM algorithm in terms of time and memory increases linearly as the dataset grows. In this study, a parallel-hybrid algorithm that combines SVM, Sequential Minimal Optimization (SMO) with Stochastic Gradient Descent (SGD) methods have been proposed to optimize the calculation of the weight costs. The performance of the proposed SVM-SMO-SGD algorithm was compared with classical SMO and Compute Unified Device Architecture (CUDA) based approaches on the well-known datasets (i.e., Diabetes, Healthcare Stroke Prediction, Adults) with 520, 5110, and 32,560 samples, respectively. According to the result...
Deep Learning-Based Prediction Models for the Detection of Vitamin D Deficiency and 25-Hydroxyvitamin D Levels Using Complete Blood Count Tests
Acı, Çiğdem | Acı, Mehmet
Vitamin D (VitD) is an essential nutrient that is critical for the well-being of both adults and children, and its deficiency is recognized as a precursor to several diseases. In previous studies, researchers have approached the problem of detecting vitamin D deficiency (VDD) as a single ”sufficient/deficient” classification problem using machine learning or statistics-based methods. The main objective of this paper is to predict a patient’s VitD status (i.e., sufficiency, insufficiency, or deficiency), severity of VDD (i.e., mild, moderate, or severe), and 25-hydroxyvitamin D (25(OH)D) level in a separate deep learning (DL)-based models. An original dataset consisting of complete blood count (CBC) tests from 907 patients, including 25(OH)D concentrations, collected from a public health la...
Evaluation of YOLOv8 Model Series with HOP for Object Detection in Complex Agriculture Domains
Bektaş, Jale
In recent years, many studies have been conducted in-depth investigating YOLO Models for object detection in the field of agriculture. For this reason, this study focused on four datasets containing different agricultural scenarios, and 20 dif-ferent trainings were carried out with the objectives of understanding the detec-tion capabilities of YOLOv8 and HPO (optimization of hyperparameters). While Weed/Crop and Pineapple datasets reached the most accurate measurements with YOLOv8n in mAP score of 0.8507 and 0.9466 respectively, the prominent model for Grapes and Pear datasets was YOLOv8l in mAP score of 0.6510 and 0.9641. This situation shows that multiple-species or in different developmental stages of a single species object YOLO training highlights YOLOv8n, while only object detection ...
Morphological and structural complexity analysis of low-resource English-Turkish language pair using neural machine translation models
Acı, Mehmet | Vuran Sarı, Nisa | Acı, Çiğdem
Neural machine translation (NMT) has achieved remarkable success in high-resource language pairs; however, its effectiveness for morphologically rich and low-resource languages like Turkish remains underexplored. As a highly agglutinative and morphologically complex language with limited high-quality parallel data, Turkish serves as a representative case for evaluating NMT systems on low-resource and linguistically challenging settings. Its structural divergence from English makes it a critical testbed for assessing tokenization strategies, attention mechanisms, and model generalizability in neural translation. This study investigates the comparative performance of two prominent NMT paradigms—the Transformer architecture, and recurrent-based sequence-to-sequence (Seq2Seq) models with atten...
Comparative Analysis of Deep Learning, Transfer Learning and Hybrid Models for Phising Detection
Vuran Sarı, Nisa | Acı, Mehmet | Sayar, Elif
2025-12-22
Enhanced pyramidal residual networks for single image super-resolution
Kahveci, Semih
Several super-resolution (SR) techniques are introduced in the literature, including traditional and machine learning-based algorithms. Especially, deep learning-based SR approaches emerge with demands for better quality images providing deeper subpixel enhancement. Dealing with the image enhancement task in the satellite images domain, a new SR method for single image SR, namely Enhanced Deep Pyramidal Residual Networks, is introduced in this study. The proposed method overcomes the potential instability problem of Enhanced Deep Residual Networks for Single Image Super-Resolution (EDSR) approach by gradually increasing the feature maps depending upon Pyramidal Residual Networks architecture. The EDSR itself is a good algorithm in the SR domain. However, it has a strict structure for incre...
Integrating a novel SRCRN network for segmentation with representative batch-mode experiments for detecting melanoma
Bektaş, Jale | Bektaş, Yasin | Kangal, Evrim Ersin
Melanoma is a type of skin cancer that tends to spread to other parts of the body and can be fatal if not detected at an early stage. This paper proposes an automated and non-invasive methodology to assist clinicians to detect melanoma. A two-stage framework was suggested in the study. In the first stage, the Resnet 50-based novel SRCRN Network was designed, which extracts high-dimensional distinctive features for skin lesion segmentation, and uses the advantage of stride regulation effectively. In the framework of SRCRN, pixel maps of different sizes were obtained by upsampling and downsampling methods between block layers, and the performance of segmentation was improved by selecting the most appropriate pixel map. In the second stage, the Resnet-50 network was used again for melanoma de...
EVALUATING THE EFFECT OF LESION SEGMENTATION ON THE DETECTION OF SKIN CANCER BY PRE-TRAINED CNN MODELS
Bektaş, Jale | Bektaş, Yasin | Kangal, Evrim Ersin
Early diagnosis of melanoma, which is considered to be one of the deadliest skin cancers, via medical imaging can significantly improve the course of the disease. However, expert assessments are subjective and open to errors due to large variations in dermoscopy images. To cope with this problem, a two-stage framework is proposed for the detection of melanoma in dermoscopic images. Firstly, by eliminating the presence of natural structures such as veins or hair and the variations in the pattern region, segmented images are obtained from raw image data with the help of pixel-wise image processing techniques. The second part of this framework is the recognition stage of the skin lesions by Pre-Trained Deep Networks (PTN). By using segmented input images, the PTN classifiers are optimized wit...
Static Analysis of Malware Detection Using Machine Learning Algorithms
Vuran, Nisa | Acı, Mehmet
2024-12-26
Automating an Encoder–Decoder Incorporated Ensemble Model: Semantic Segmentation Workflow on Low-Contrast Underwater Images
Bektaş, Jale
Numerous methods have been proposed for semantic segmentation and the state-of-the-art part is likely to be incorporated by deep learning-based methods which show a salient performance. This study addresses the challenge of semantic segmentation in low-contrast imbalanced under water images. Moreover, it employs nine model fusions as a downstream workflow task using encoder–decoder architectures with Dice Loss and Focal Loss training focusing on the imbalance data. Afterwards, the most effective two encoder–decoder fusion models, Res34+Unet and VGG19+FPN, by 0.592%, 0.590% mIoU on average and by 0.510%, 0.491% F1-score yielded better performance, respectively, than other models. Using a weight-optimization algorithm, the ensemble model with recreated IoU results improves the accuracy for b...
Underwater Image Enhancement Multi-Exposure Fusion Using Unsupervised Deep Learning
Kahveci, Semih | Avaroğlu, Erdinç
2025-12-22
TARIM VE ORMANCILIK YÖNETİMİNDE NESNE TESPİT ALGORİTMALARININ KULLANIM ALANLARINA GENEL BİR BAKIŞ
Akkurt, Ramazan | Kahveci, Semih
Nesne tespit algoritmaları literatüre kazandırıldıktan kısa bir süre sonra birçok farklı alanda kullanımı yaygınlaşmıştır. Tarım ve ormancılık yönetimi de nesne tespit algoritmalarının kullanıldığı alanların en başında gelmektedir. Tarım ve ormancılık yönetimindeki birçok problemlere uyarlanan nesne tespiti algoritmaları, bu alana ciddi katkılar sağlamaktadır. Ağaç sınıflandırma, ölü ağaç örtüsü tespiti, verimli üretim, mahsul izleme, hastalık tespiti vb. birçok konu başlığında tarım ve ormancılık faaliyetlerinde yaygın şekilde kullanılmaktadır. Nesne tespit algoritmalarının sunduğu avantajlar bu alandaki insan hatalarının minimum seviyelere indirildiği, maliyetlerin azaldığı ve mahsul verimliliğin ciddi artışlar sağladığı literatürde yapılan birçok çalışma ile doğrulanmıştır. Bu çalışma k...