Computer Science
Permanent URI for this collectionhttps://hdl.handle.net/10679/9120
Browse
Recent Submissions
Conference paperPublication Metadata only International roaming traffic optimization with call quality(SciTePress, 2019) Şahin, Ahmet; Demirel, Kenan Cem; Albey, Erinç; Gürsun, Gonca; Industrial Engineering; ALBEY, Erinç; Şahin, Ahmet; Demirel, Kenan Cem; Gürsun, GoncaIn this study we focus on a Steering International Roaming Traffic (SIRT) problem with single service that concerns a telecommunication’s operators’ agreements with other operators in order to enable subscribers access services, without interruption, when they are out of operators’ coverage area. In these agreements, a subscriber’s call from abroad is steered to partner operator. The decision for which each call will be forwarded to the partner is based on the user’s location (country/city), price of the partner operator for that location and the service quality of partner operator. We develop an optimization model that considers agreement constraints and quality requirements while satisfying subscribers demand over a predetermined time interval. We test the performance of the proposed approach using different execution policies such as running the model once and fixing the roaming decisions over the planning interval or dynamically updating the decisions using a rolling horizon approach. We present a rigorous trade off analysis that aims to help the decision maker in assessing the relative importance of cost, quality and ease of implementation. Our results show that steering cost is decreased by approximately 25% and operator mistakes are avoided with the developed optimization model while the quality of the steered calls is kept above the base quality level.Conference paperPublication Metadata only Uncertainty-aware deep classifiers using generative models(Association for the Advancement of Artificial Intelligence, 2020) Şensoy, Murat; Kaplan, L.; Cerutti, F.; Saleki, Maryam; Computer Science; ŞENSOY, Murat; Saleki, MaryamDeep neural networks are often ignorant about what they do not know and overconfident when they make uninformed predictions. Some recent approaches quantify classification uncertainty directly by training the model to output high uncertainty for the data samples close to class boundaries or from the outside of the training distribution. These approaches use an auxiliary data set during training to represent out-of-distribution samples. However, selection or creation of such an auxiliary data set is non-trivial, especially for high dimensional data such as images. In this work we develop a novel neural network model that is able to express both aleatoric and epistemic uncertainty to distinguish decision boundary and out-of-distribution regions of the feature space. To this end, variational autoencoders and generative adversarial networks are incorporated to automatically generate out-of-distribution exemplars for training. Through extensive analysis, we demonstrate that the proposed approach provides better estimates of uncertainty for in- and out-of-distribution samples, and adversarial examples on well-known data sets against state-of-the-art approaches including recent Bayesian approaches for neural networks and anomaly detection methods.Conference paperPublication Metadata only A benchmark for inpainting of clothing images with irregular holes(Springer, 2020) Kınlı, Osman Furkan; Özcan, Barış; Kıraç, Mustafa Furkan; Computer Science; KINLI, Osman Furkan; KIRAÇ, Mustafa Furkan; Özcan, BarışFashion image understanding is an active research field with a large number of practical applications for the industry. Despite its practical impacts on intelligent fashion analysis systems, clothing image inpainting has not been extensively examined yet. For that matter, we present an extensive benchmark of clothing image inpainting on well-known fashion datasets. Furthermore, we introduce the use of a dilated version of partial convolutions, which efficiently derive the mask update step, and empirically show that the proposed method reduces the required number of layers to form fully-transparent masks. Experiments show that dilated partial convolutions (DPConv) improve the quantitative inpainting performance when compared to the other inpainting strategies, especially it performs better when the mask size is 20% or more of the image.ArticlePublication Metadata only Centrality and scalability analysis on distributed graph of large-scale e-mail dataset for digital forensics(IEEE, 2020-12-10) Ozcan, S.; Astekin, Merve; Shashidhar, N. K.; Zhou, B.; Astekin, MerveToday's digital forensics software tools mostly do not offer automatic analysis methods to reveal evidences among huge amounts of digital files within hard disk images. It is important that finding evidence in digital and cyber forensics investigations as soon as possible by examining hard disk images. E-mails constitute a rich source of information in hard disk images, and they are the most possible data source to obtain an evidence. The analyzers search e-mail files by manually or using traditional methods in order to find an evidence. However, this operation could take a long time due to the size of the e-mail data which can contain a huge number of files and a huge volume of data. This study introduces an end-to-end distributed graph analysis framework for large-scale digital forensic datasets, and evaluates the accuracy of the centrality algorithms and the scalability of the proposed framework in terms of running time performance. The framework is comprised of specific processes to perform pre-processing, graph building, and algorithm activities. An architecture is introduced based on distributed big data techniques. Three different centrality algorithms are implemented to analyze the accuracy of our framework. Further, three implementations are provided to demonstrate the running time performance of our framework. Experiments are performed on Enron e-mail dataset to analyze the centrality algorithms, to evaluate the performance of the framework, and to compare the running times between the traditional approach and our approach. Moreover, the running time performance of the framework is evaluated under various parallelization level. The accuracy of the results is also evaluated and compared between the centrality algorithms. The comparison shows that some certain algorithms provide more accurate results and it is possible to improve the running time by orders of magnitude utilizing our end-to-end distributed graph analysis approach.Conference paperPublication Metadata only Exploring scaling efficiency of intel loihi neuromorphic processor(IEEE, 2023) Uludağ, Recep Buğra; Çaǧdaş, S.; Işler, Y. S.; Şengör, N. S.; Aktürk, İsmail; Computer Science; AKTÜRK, Ismail; Uludağ, Recep BuğraIn this paper, we focus on examining how scaling efficiency evolves in winner-take-all (WTA) network models on Intel Loihi neuromorphic processor, as network-related features such as network size, neuron type, and connectivity scheme change. By analyzing these relationships, our study aims to shed light on the intricate interplay between SNN features and the efficiency of neuromorphic systems as they scale up. The findings presented in this paper are expected to enhance the comprehension of scaling efficiency in neuromorphic hardware, providing valuable insights for researchers and developers in optimizing the performance of large-scale SNNs on neuromorphic architectures.Conference paperPublication Metadata only Effects of agent's embodiment in human-agent negotiations(ACM, 2023-09-19) Çakan, Umut; Keskin, Mehmet Onur; Aydoǧan, Reyhan; Computer Science; AYDOĞAN, Reyhan; Çakan, Umut; Keskin, Mehmet OnurHuman-agent negotiation has recently attracted researchers’ attention due to its complex nature and potential usage in daily life scenarios. While designing intelligent negotiating agents, they mainly focus on the interaction protocol (i.e., what to exchange and how) and strategy (i.e., how to generate offers and when to accept). Apart from these components, the embodiment may implicitly influence the negotiation process and outcome. The perception of a physically embodied agent might differ from the virtually embodied one; thus, it might influence human negotiators’ decisions and responses. Accordingly, this work empirically studies the effect of physical and virtual embodiment in human-agent negotiations. We designed and conducted experiments where human participants negotiate with a humanoid robot in one setting, whereas they negotiate with a virtually embodied replica of that robot in another setting. The experimental results showed that social welfare was statistically significantly higher when the negotiation was held with a virtually embodied robot rather than a physical robot. Human participants took the negotiation more seriously against physically embodied agents and made more collaborative moves in the virtual setting. Furthermore, their survey responses indicate that participants perceived our robot as more humanlike when it is physically embodied.ReviewPublication Open Access Deep learning-based expressive speech synthesis: a systematic review of approaches, challenges, and resources(Springer, 2024-02-12) Barakat, Huda Mohammed Mohammed; Turk, O.; Demiroğlu, Cenk; Electrical & Electronics Engineering; DEMİROĞLU, Cenk; Barakat, Huda Mohammed MohammedSpeech synthesis has made significant strides thanks to the transition from machine learning to deep learning models. Contemporary text-to-speech (TTS) models possess the capability to generate speech of exceptionally high quality, closely mimicking human speech. Nevertheless, given the wide array of applications now employing TTS models, mere high-quality speech generation is no longer sufficient. Present-day TTS models must also excel at producing expressive speech that can convey various speaking styles and emotions, akin to human speech. Consequently, researchers have concentrated their efforts on developing more efficient models for expressive speech synthesis in recent years. This paper presents a systematic review of the literature on expressive speech synthesis models published within the last 5 years, with a particular emphasis on approaches based on deep learning. We offer a comprehensive classification scheme for these models and provide concise descriptions of models falling into each category. Additionally, we summarize the principal challenges encountered in this research domain and outline the strategies employed to tackle these challenges as documented in the literature. In the Section 8, we pinpoint some research gaps in this field that necessitate further exploration. Our objective with this work is to give an all-encompassing overview of this hot research area to offer guidance to interested researchers and future endeavors in this field.ArticlePublication Open Access Towards interactive explanation-based nutrition virtual coaching systems(Springer, 2024-01) Buzcu, Berk; Tessa, M.; Tchappi, I.; Najjar, A.; Hulstijn, J.; Calvaresi, D.; Aydoğan, Reyhan; Computer Science; AYDOĞAN, Reyhan; Buzcu, BerkThe awareness about healthy lifestyles is increasing, opening to personalized intelligent health coaching applications. A demand for more than mere suggestions and mechanistic interactions has driven attention to nutrition virtual coaching systems (NVC) as a bridge between human–machine interaction and recommender, informative, persuasive, and argumentation systems. NVC can rely on data-driven opaque mechanisms. Therefore, it is crucial to enable NVC to explain their doing (i.e., engaging the user in discussions (via arguments) about dietary solutions/alternatives). By doing so, transparency, user acceptance, and engagement are expected to be boosted. This study focuses on NVC agents generating personalized food recommendations based on user-specific factors such as allergies, eating habits, lifestyles, and ingredient preferences. In particular, we propose a user-agent negotiation process entailing run-time feedback mechanisms to react to both recommendations and related explanations. Lastly, the study presents the findings obtained by the experiments conducted with multi-background participants to evaluate the acceptability and effectiveness of the proposed system. The results indicate that most participants value the opportunity to provide feedback and receive explanations for recommendations. Additionally, the users are fond of receiving information tailored to their needs. Furthermore, our interactive recommendation system performed better than the corresponding traditional recommendation system in terms of effectiveness regarding the number of agreements and rounds.Conference paperPublication Metadata only Context based echo state networks for robot movement primitives(IEEE, 2023) Amirshirzad, Negin; Asada, M.; Öztop, Erhan; Computer Science; ÖZTOP, Erhan; Amirshirzad, NeginReservoir Computing, in particular Echo State Networks (ESNs) offer a lightweight solution for time series representation and prediction. An ESN is based on a discrete time random dynamical system that is used to output a desired time series with the application of a learned linear readout weight vector. The simplicity of the learning suggests that an ESN can be used as a lightweight alternative for movement primitive representation in robotics. In this study, we explore this possibility and develop Context-based Echo State Networks (CESNs), and demonstrate their applicability to robot movement generation. The CESNs are designed for generating joint or Cartesian trajectories based on a user definable context input. The context modulates the dynamics represented by the ESN involved. The linear read-out weights then can pick up the context-dependent dynamics for generating different movement patterns for different contexts. To achieve robust movement execution and generalization over unseen contexts, we introduce a novel data augmentation mechanism for ESN training. We show the effectiveness of our approach in a learning from demonstration setting. To be concrete, we teach the robot reaching and obstacle avoidance tasks in simulation and in real-world, which shows that the developed system, CESN provides a lightweight movement primitive representation system that facilitate robust task execution with generalization ability for unseen seen contexts, including extrapolated ones.ArticlePublication Metadata only EPIoT: Enhanced privacy preservation based blockchain mechanism for internet-of-things(Elsevier, 2024-01) Kashif, Muhammad; Çakmakçı, Kübra Kalkan; Computer Science; ÇAKMAKCİ, Kübra Kalkan; Kashif, MuhammadWith the increasing popularity of the Internet of things (IoT) and giving the end users the opportunity of collecting and analyzing the data by these IoT devices give rise to ultimate privacy concern and is attracting significant attention nowadays. These IoT devices may contain highly sensitive data and data sharing processes which may lead to security and privacy concerns. To surmount these issues, the interaction of IoT with blockchain for a secure transaction is accepted as a candidate solution. However, the innate behavior of blockchain containing complex mathematical proofs and consensus protocol requires high computational power making it less favorable for IoT devices to be connected with. Motivated by a private by-design framework and emphasizing greater control and setting of privacy preferences by the data owner, this paper complements our previous work on privacy preservation in IoT networks. In this paper, we design and propound a complete blockchain-based privacy-preserving framework by deploying service-oriented layers concepts and low computation cryptography, and a less complex consensus protocol to address the privacy concern. Moreover, this paper will unravel the complete end-to-end architecture of IoT-based blockchain purposely build for secure transactions in IoT networks. Security analysis is conducted using AVISPA tool to show that the proposed algorithms attain the desired security goals. This is followed by extensive simulation experiments and ultimate output results cultivating it much favorably for the deployment of IoT applications in real life.Conference paperPublication Metadata only Towards test automation for certification tests in the banking domain(IEEE, 2023) Elakas, A.; Tarlan, Ozan; Safak, I.; Çakmakçı, Kübra Kalkan; Sözer, Hasan; Computer Science; SÖZER, Hasan; ÇAKMAKCİ, Kübra Kalkan; Tarlan, OzanSoftware systems in the banking domain are business-critical applications that provide financial services. These systems are subject to rigorous certification tests, which are performed manually, and take weeks to complete. In this paper, we suggest that automation of the certificate tests are possible and it will save a considerable amount of time. A certification testing operation which can take a few weeks can be reduced to a few seconds. Firstly, we review the existing test activities to identify the ones that can be automated and introduce a prototype tool for automating some of the tests used for certification. We focus on rules that are verified by analyzing the banking infrastructure. Our tool takes the network topology of the banking infrastructure as input and verifies a subset of these rules. The tool can be extended with additional rules in order to reduce the effort for certification tests. In addition to this tool, we introduce softwaredefined network-based tests to automatically verify compliance with the rules by checking the firewall constraints and host connections. In particular, we focus on a security certification standard named Payment Card Industry Data Security Standard. This certification aims to reduce the risk of data breaches in cardholder data by ensuring industry standard practices for payment card transactions. Our tool offers effort reduction in auditing through automation. It supports continuous auditing and network security enhancement processes.Conference paperPublication Metadata only Deterministic neural illumination mapping for efficient auto-white balance correction(IEEE, 2023) Kınlı, Osman Furkan; Yılmaz, Doğa; Özcan, Barış; Kıraç, Mustafa Furkan; Computer Science; KINLI, Osman Furkan; KIRAÇ, Mustafa Furkan; Yılmaz, Doğa; Özcan, BarışAuto-white balance (AWB) correction is a critical operation in image signal processors for accurate and consistent color correction across various illumination scenarios. This paper presents a novel and efficient AWB correction method that achieves at least 35 times faster processing with equivalent or superior performance on high-resolution images for the current state-of-the-art methods. Inspired by deterministic color style transfer, our approach introduces deterministic illumination color mapping, leveraging learnable projection matrices for both canonical illumination form and AWB-corrected output. It involves feeding high-resolution images and corresponding latent representations into a mapping module to derive a canonical form, followed by another mapping module that maps the pixel values to those for the corrected version. This strategy is designed as resolution-agnostic and also enables seamless integration of any pre-trained AWB network as the backbone. Experimental results confirm the effectiveness of our approach, revealing significant performance improvements and reduced time complexity compared to state-of-the-art methods. Our method provides an efficient deep learning-based AWB correction solution, promising real-time, high-quality color correction for digital imaging applications.ArticlePublication Metadata only Applying deep learning models to twitter data to detect airport service quality(Elsevier, 2021-03) Barakat, Huda Mohammed Mohammed; Yeniterzi, R.; Martin-Domingo, Luis; Aviation Management; DOMINGO, Luıs Martın; Barakat, Huda Mohammed MohammedMeasuring airport service quality (ASQ) is an important process for identifying shortages and suggesting improvements that guide management decisions. This research, introduces a general framework for measuring ASQ using passengers’ tweets about airports. The proposed framework considers tweets in any language, not just in English, to support ASQ evaluation in non-speaking English countries where passengers communicate with other languages. Accordingly, this work uses a large dataset that includes tweets in two languages (English and Arabic) and from four airports. Additionally, to extract passenger evaluations from tweets, our framework applies two different deep learning models (CNN and LSTM) and compares their results. The two models are trained with both general data and data from the aviation domain in order to clarify the effect of data type on model performance. Results show that better performance is achieved with the LSTM model when trained with domain specific data. This study has clear implications for researchers and airport managers aiming to use alternative methods to measure ASQ.ArticlePublication Metadata only Provenance aware run-time verification of things for self-healing Internet of Things applications(Wiley, 2019-02-10) Aktas, M. S.; Astekin, Merve; Astekin, MerveWe propose a run-time verification mechanism of things for self-healing capability in the Internet of Things domain. We discuss the software architecture of the proposed verification mechanism and its prototype implementations. To identify faulty running behavior of things, we utilize a complex event processing technique by applying rule-based pattern detection on the events generated real time. For events, we use a descriptor metadata of the measurements (such as CPU usage, memory usage, and bandwidth usage) taken from Internet of Things devices. To understand the usability and effectiveness of the proposed mechanism, we developed prototype applications using different event processing platforms. We test the prototype implementations for performance and scalability under increasing message rates. The results are promising because the processing overhead of the proposed verification mechanism is negligible.