Research Outputs
Permanent URI for this communityhttps://hdl.handle.net/20.500.14288/2
Browse
16 results
Search Results
Publication Metadata only Complaint detection and classification of customer reviews(IEEE, 2021) Bayrak, Ahmet Tuğrul; Yıldız, Eray; Özbek, Eyüp Erkan; Türker, Bekir Berker; PhD Student; Graduate School of Sciences and Engineering; N/AIn a world where competition and technology usage increase consistently, customer satisfaction has become important for companies. In this study, the customer reviews, obtained from the results of the surveys that are made via different channels, are analyzed and when a problem is detected, a quick solution is aimed. For the complaint detection and classification on the customer reviews process, long short-term memory, which is a recurrent neural network, is applied. A data set from the tourism industry is labelled to carry out the proposed method. The results retrieved on performing the method on the data, which is relatively larger than the similar works in literature, are acceptable and the proposed model works in real-time.Publication Metadata only Devoloping affordable tangible programming education applications using mobile vision(IEEE, 2021) N/A; Department of Computer Engineering; Sabuncuoğlu, Alpay; Sezgin, Tevfik Metin; PhD Student; Faculty Member; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 18632Programming education has become an essential part of the primary and secondary school curriculum. Two main programming environment modalities, web-based visual programming applications and electronic cards, have become popular in curricular activities. Limiting the programming activities around these programming environments restricts education accessibility in socio-economically disadvantaged regions due to the need for an individual computer per student, lack of decent infrastructure, and high electronics prices. Effective, shared use of smartphones and tablets in programming education can provide equal opportunities. in this scenario, students can code simple drawings and animations using their own materials as tangible programming blocks by employing a single shared phone in the classroom as an interpreter. This article explains our development process of a new tangible programming environment which increases the accessibility of education. We discuss effective inclassroom use of image/text processing practices and transfer learning methods on smartphones.Publication Metadata only Distributed management of service-enabled flow-paths across multiple SDN domains(IEEE, 2015) Civanlar, Seyhan; Lokman, Erhan; Kaytaz, Bülent; Department of Electrical and Electronics Engineering; Tekalp, Ahmet Murat; Faculty Member; Department of Electrical and Electronics Engineering; College of Engineering; 26207Future Internet may be comprised of interconnected multiple software-defined networks (SDN), where each domain is administered by a different controller or control plane. Provisioning of an end-to-end flow-path across such a network with specific service levels requires collaboration between domain controllers across control planes. A service-enabled flow-path shall be quality of service (QoS) enabled, reliable and/or secure, which requires a transport path with certain quantitative service level requirements such as high throughput, low packet loss, or high availability. Each SDN controller can autonomously determine such end-to-end flow-paths when all other SDN controllers periodically advertise to other controllers its available service-enabled paths. Doing so, each SDN controller is presented with several service-enabled path alternatives, crossing other domains, to choose from. We propose a multi-domain SDN controller design, wherein each SDN controller shares its network's "summarized" topology of service-enabled paths with other SDN controllers, such that all domains (controllers) have real-time autonomous decision making capability for end-to-end flow-path selection. We also describe how an SDN controller can reserve and release a flow traversing other SDN domains.Publication Metadata only EdgeKV: distributed key-value store for the network edge(IEEE, 2020) Al Oqily, Ibrahim; Aloqaily, Moayad; N/A; Department of Computer Engineering; Sonbol, Karim; Özkasap, Öznur; Master Student; Faculty Member; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 113507With improvements in computation and storage resources, data access through the network becomes the bottleneck for several cloud applications. Even with high-speed networks, the high latency of the cloud access makes it unfeasible or unfavourable for latency-sensitive applications such as autonomous driving, smart factories, and video streaming. Edge computing provides a solution by utilizing the network edge resources that are closer to the end users. Nevertheless, it is a non-trivial task to design a large-scale edge-capable system that is stable, fault-tolerant, and efficient [1]. In this paper, present the design of EdgeKV: a novel general-purpose distributed key-value store for the network edge. We demonstrate the features of EdgeKV for achieving high efficiency and scalability while providing flexibility, ease of use, and data privacy. We evaluated our prototype on the Grid'5000 framework with multiple realistic Yahoo! Cloud Serving Benchmark (YCSB) workloads. Our initial results show that EdgeKV achieves 72% higher throughput and 47% lower latency on average than centralized cloud storage, for read-dominated workloads.Publication Metadata only Effects of transmit beamforming on the capacity of multi-hop MIMO relay channels(IEEE, 2007) Yılmaz, Erhan; Department of Electrical and Electronics Engineering; Sunay, Mehmet Oğuz; Faculty Member; Department of Electrical and Electronics Engineering; College of Engineering; N/AIn this paper we investigate multi-hop fading relay channels where the source, the destination and the multiple relay nodes are all equipped with multiple antennas. We study the ergodic capacities of multiple relay networks based on Amplify-and-Forward (AF) and Decode-and-Forward (DF) relaying modes. We examine multi-user beamforming (MU-BF), where each data stream is assumed to be matched to a specified relay node, based on the conventional eigen-mode transmission for both modes, and derive ergodic capacity expressions. We also examine the impact of the number of selected relay nodes on the network capacity both for modes of relaying. We show that by using MU-BF at the source node and a maximum number of relay nodes selected for cooperation, the network gains from multiplexing, while beyond that number of cooperating relay nodes, it only gains from relay selection.Publication Metadata only Employing 60 GHz ISM band for 5G wireless communications(IEEE, 2014) Fadel, Etimad; N/A; Department of Electrical and Electronics Engineering; Yılmaz, Türker; Akan, Özgür Barış; PhD Student; Faculty Member; Department of Electrical and Electronics Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 6647Wireless data traffic is continuously increasing due to the steady rise in both connected device number and traffic per device. Wireless networks, traditionally confined below 6 giga-hertz, are getting clogged and unable to satisfy the ever-increasing demands of its users. Already aware of this, telecommunications industry and academia have been working on solutions. One of the main methods for throughput increase is operation bandwidth expansion; however, sufficient spectrum is not available within the conventional frequencies. Following various considerations, 60 GHz industrial, scientific and medical radio band has been selected as the new spectrum to be utilized and wireless personal and local area network standards for the band are already completed. In line with the stated developments, this paper proposes the use of 60 GHz band for the fifth generation (5G) communication systems. After very briefly setting the scene of the current wireless communication networks, the physical layer properties of the 60 GHz band are presented. A representative indoor simulation between the fourth generation and proposed 5G cases is set and performed. The results are assessed and compared before concluding the paper.Publication Metadata only Few-shot learning for segmentation of yeast cell microscopy images(IEEE, 2021) Alkan, Muhammet; Kiraz, Berna; Eren, Furkan; N/A; Department of Physics; Uysallı, Yiğit; Kiraz, Alper; PhD Student; Faculty Member; Department of Physics; Graduate School of Sciences and Engineering; College of Sciences; N/A; 22542Cell segmentation from microscopic images can be performed using deep neural networks or image processing techniques. In addition to their inherent difficulties, these techniques come together with the requirement of feeding the neural network with a large number of image samples in order to obtain a good result. However, this is not sustainable in terms of collecting and labeling microscopic images and represents a costly and time-consuming solution for every new microscopic image and cell type. Instead, fine-tuning can be employed by taking advantage of the adaptation ability of a model trained using meta-learning algorithms. In this way, while more general and better results can be obtained with fewer samples, the training process does not start from scratch for each new cell type or data set. In this article, microscopic images of yeast cells were recorded and analyzed using Reptile algorithm. After fine-tuning with a small number of samples, an average success rate of 81 % IoU (Intersection over Union) was obtained on the test pictures in addition to the model accuracy reaching up to 87%.Publication Metadata only Lightwave circuit elements based on microsphere resonators and meandering waveguides(IEEE, 2016) Department of Physics; Serpengüzel, Ali; Faculty Member; Department of Physics; College of Sciences; 27855The microspheres, with their high quality factor morphology dependent resonances (MDRs), are ideal optical resonators for three dimensional volumetric lightwave circuits. The microsphere leads itself to various lightwave circuit element applications such as channel dropping filters [1], tunable filters [2], optical modulators [3], and dynamic tuners [4]. So far we have realized these applications using silicon spheres coupled with optical fiber half couplers manufactured from single mode optical fibers. Biosensing might also be realized with these silicon microsphere resonators [5]. On the integrated optics side, we recently introduced distributed feedback (DFB) meandering waveguides as novel integrated optical elements [6]. We analyzed silicon DFB meandering waveguides, which can exhibit a variety of spectral responses such as coupled resonator induced transparency (CRIT) filter, Fano resonator, hitless filter, Lorentzian filter, Rabi splitter, self-coupled optical waveguide (SCOW), and tunable power divider. In this talk, we will focus on the properties of silicon spherical resonators and distributed feedback (DFB) meandering waveguides and their potential as lightwave components.Publication Metadata only Millimeter-wave communications for 5G wireless networks(Crc Press-Taylor and Francis Group, 2016) N/A; Department of Electrical and Electronics Engineering; Yılmaz, Türker; Akan, Özgür Barış; PhD Student; Faculty Member; Department of Electrical and Electronics Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 6647N/APublication Metadata only Minimum length scheduling for power constrained harvest-then-transmit communication networks(IEEE, 2019) N/A; N/A; Department of Electrical and Electronics Engineering; Salık, Elif Dilek; Önalan, Aysun Gurur; Ergen, Sinem Çöleri; PhD Student; PhD Student; Faculty Member; Department of Electrical and Electronics Engineering; Graduate School of Sciences and Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; N/A; 7211We consider a wireless powered, harvest-then-transmit communication network, which consists of a single antenna, energy and information access point (aP) and multiple, single antenna, batteryless users with energy harvesting capabilities. at the beginning of a time frame, the aP broadcasts energy in the downlink to the users. then, users transmit their data to the aP in the uplink, using their harvested energy. We formulate the optimization problem with the objective of minimizing the total schedule length, subject to the constraints on the minimum amount of data to be sent to the aP, and unlike previous studies, the maximum transmit power for the information transmission. This problem is nonlinear and non-convex. the solution is based on bi-level optimization, consisting of optimizing the transmit power allocation of the nodes for a given energy harvesting time and searching over harvesting time allocation. We also propose a heuristic algorithm in which we incorporate the optimal solution of a single user network. Simulation results demonstrate that under appropriate network conditions, our proposed algorithms provide close-to-optimal results with a reasonable run time compared to a previously proposed time minimization algorithm that does not integrate the uplink power constraint.