Browsing by Author "Saran, Ayse Nurdan"

Now showing 1 - 6 of 6

Enhancing Session-Based Trip Recommendations Using Matrix Factorization: a Study on Algorithm Efficiency and Resource Utilization
(Springer, 2025) Mat, Abdullah Ugur; Saran, Ayse Nurdan; 06.01. Bilgisayar Mühendisliği; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
As the impact and usefulness of recommendation systems continue to grow, their importance becomes more and more pronounced. Therefore, it is crucial to design and implement recommendation systems that are both efficient and highly accurate to meet the increasing demands and expectations. This study focuses on a model awarded first place in a travel forecasting recommendation system competition. This study aims to enhance matrix factorization-based recommender systems by conducting a comprehensive analysis of various factors. This includes examining the effects of resource utilization and recurrent neural network (RNN) algorithms on session-based factorization, as well as evaluating the influence of embeddings and optimization techniques concerning their efficiency and accuracy. The gated recurrent unit (GRU) algorithm has produced more accurate results for reduced datasets than long short-term memory (LSTM). Some modifications have been made on the embedding layers, and the results have been observed. In addition, the model's optimizer is changed, and the performance of different optimizers is evaluated. While random reduction of the dataset has led to a decrease in the success rate, methodical reduction has significantly increased the success rate. The highest and most reliable success rate (0.6654) was achieved by applying the selection method, which reduced the dataset to 1 M records from 1.5 M records. Optimizers have shown a wide range of effects on hardware.
Citation - WoS: 2
Citation - Scopus: 3
On Time-Memory Trade-Offs for Password Hashing Schemes
(Frontiers Media Sa, 2024) Saran, Ayse Nurdan; 20868; 06.01. Bilgisayar Mühendisliği; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
A password hashing algorithm is a cryptographic method that transforms passwords into a secure and irreversible format. It is used not only for authentication purposes but also for key derivation mechanisms. The primary purpose of password hashing is to enhance the security of user credentials by preventing the exposure of plaintext passwords in the event of a data breach. As a key derivation function, password hashing aims to derive secret keys from a master key, password, or passphrase using a pseudorandom function. This review focuses on the design and analysis of time-memory trade-off (TMTO) attacks on recent password hashing algorithms. This review presents a comprehensive survey of TMTO attacks and recent studies on password hashing for authentication by examining the literature. The study provides valuable insights and strategies for safely navigating transitions, emphasizing the importance of a systematic approach and thorough testing to mitigate risk. The purpose of this paper is to provide guidance to developers and administrators on how to update cryptographic practices in response to evolving security standards and threats.
Parallelization of Sparsity-Driven Change Detection Method
(Ieee, 2017) Ozgur, Atilla; Saran, Ayse Nurdan; Nar, Fatih; 02.02. Matematik; 06.01. Bilgisayar Mühendisliği; 02. Fen-Edebiyat Fakültesi; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
In this study, Sparsity-driven Change Detection (SDCD) method, which has been proposed for detecting changes in multitemporal synthetic aperture radar (SAR) images, is parallelized to reduce the execution time. Parallelization of the SDCD is realized using OpenMP on CPU and CUDA on GPU. Execution speed of the parallelized SDCD is shown on real-world SAR images. Our experimental results show that the computation time of the parallel implementation brings significant speed-ups.
Citation - WoS: 8
Citation - Scopus: 15
Perlin Random Erasing for Data Augmentation
(Ieee, 2021) Saran, Ayse Nurdan; Saran, Murat; Nar, Fatih; 17753; 20868; 06.01. Bilgisayar Mühendisliği; 02.02. Matematik; 02. Fen-Edebiyat Fakültesi; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
In the last decade, Deep Learning is applied in a wide range of problems with tremendous success. Large data, increased computational resources, and theoretical improvements are main reasons for this success. As the dataset grows, the real-world is better represented, allows developing a model that can generalize. However, creating a labeled dataset is expensive, time-consuming, or sometimes even challenging. Therefore, researchers proposed data augmentation methods to increase dataset size by creating variations of the existing data. This study proposes an extension to Random Erasing data augmentation method by introducing smoothness. The proposed method provides better performance compared to Random Erasing data augmentation method, which is shown using a transfer learning scenario on the UC Merced Land-use image dataset.
Citation - WoS: 13
Citation - Scopus: 15
Sparsity-Driven Change Detection in Multitemporal Sar Images
(Ieee-inst Electrical Electronics Engineers inc, 2016) Saran, Ayse Nurdan; Nar, Fatih; Ozgur, Atilla; 252953; 20868; 02.02. Matematik; 06.01. Bilgisayar Mühendisliği; 02. Fen-Edebiyat Fakültesi; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
In this letter, a method for detecting changes in multitemporal synthetic aperture radar (SAR) images by minimizing a novel cost function is proposed. This cost function is constructed with log-ratio-based data fidelity terms and an l(1)-norm-based total variation (TV) regularization term. Log-ratio terms model the changes between the two SAR images where the TV regularization term imposes smoothness on these changes in a sparse manner such that fine details are extracted while effects like speckle noise are reduced. The proposed method, sparsity-driven change detection (SDCD), employs accurate approximation techniques for the minimization of the cost function since data fidelity terms are not convex and the employed l(1)-norm TV regularization term is not differentiable. The performance of the SDCD is shown on real-world SAR images obtained from various SAR sensors.
Unified Lf-Norm Robust Fitting for Linear Models
(Institute of Electrical and Electronics Engineers Inc., 2025) Nar, Fatih; Saran, Murat; Saran, Ayse Nurdan; Şen, Baha; 02.02. Matematik; 06.01. Bilgisayar Mühendisliği; 02. Fen-Edebiyat Fakültesi; 06. Mühendislik Fakültesi; 01. Çankaya Üniversitesi
In statistical learning, accurately estimating model parameters is crucial for reliable predictions. Managing residuals, the differences between observed and predicted values, is a key challenge. In regression, the residual penalty choice strongly affects model performance. The L2-norm penalty aligns with the least-squares approach, while the L1-norm provides robust fitting by minimizing the influence of outliers. To generalize models, the weights can be regularized using either the L2-norm or L1-norm, corresponding to Ridge and LASSO regularization, respectively. Many methods have been developed to penalize residuals and model weights, resulting in diverse cost functions optimized by specific numerical solvers. In this study, we propose the smooth Lf-norm, a quasi-norm, as a unified framework for penalizing both residuals and model weights in linear models. Our efficient and robust numerical minimization scheme ensures fast and accurate fitting by minimizing our novel cost function. © 2025 Elsevier B.V., All rights reserved.