Publications with Fulltext
Permanent URI for this collectionhttps://hdl.handle.net/20.500.14288/6
Browse
83 results
Search Results
Publication Open Access Evolutionary multiobjective feature selection for sentiment analysis(Institute of Electrical and Electronics Engineers (IEEE), 2021) Pelin Angın; Deniz, Ayça; Department of International Relations; Angın, Merih; Faculty Member; Department of International Relations; College of Administrative Sciences and Economics; 308500Sentiment analysis is one of the prominent research areas in data mining and knowledge discovery, which has proven to be an effective technique for monitoring public opinion. The big data era with a high volume of data generated by a variety of sources has provided enhanced opportunities for utilizing sentiment analysis in various domains. In order to take best advantage of the high volume of data for accurate sentiment analysis, it is essential to clean the data before the analysis, as irrelevant or redundant data will hinder extracting valuable information. In this paper, we propose a hybrid feature selection algorithm to improve the performance of sentiment analysis tasks. Our proposed sentiment analysis approach builds a binary classification model based on two feature selection techniques: an entropy-based metric and an evolutionary algorithm. We have performed comprehensive experiments in two different domains using a benchmark dataset, Stanford Sentiment Treebank, and a real-world dataset we have created based on World Health Organization (WHO) public speeches regarding COVID-19. The proposed feature selection model is shown to achieve significant performance improvements in both datasets, increasing classification accuracy for all utilized machine learning and text representation technique combinations. Moreover, it achieves over 70% reduction in feature size, which provides efficiency in computation time and space.Publication Open Access Power performance of a continuous-wave Cr2+:ZnSe laser at 2.4 7 ?m(Optica Publishing Group, 2000) Pollock, C.R.; Department of Physics; Sennaroğlu, Alphan; Konca, Ali Özgün; Faculty Member; Undergraduate Student; Department of Physics; College of Sciences; 23851; N/AContinuous-wave power performance of a Cr2+:znSe laser was investigated at 2.474 ?m. End pumped by a 1.583-?m NaCl:OH- laser, the resonator with a 3% transmitting output coupler produced as high as 250 mW of output power with a slope efficiency of 24.2%. Analysis of the laser efficiency data shows that the magnitude of the excited-state absorption cross section is less than 5% of the emission cross section in agreement with spectroscopic results. Numerical calculations further predict the optimum crystal length and absorption coefficient to be 2.5 cm and 0.49 cm-1, respectively, for continuous-wave operation.Publication Open Access Minimum energy coding for wireless nanosensor networks(Institute of Electrical and Electronics Engineers (IEEE), 2012) Kocaoğlu, Murat; Akan, Özgür Barış; Faculty Member; College of EngineeringWireless nanosensor networks (WNSNs), which are collections of nanosensors with communication units, can be used for sensing and data collection with extremely high resolution and low power consumption for various applications. In order to realize WNSNs, it is essential to develop energy-efficient communication techniques, since nanonodes are severely energy-constrained. In this paper, a novel minimum energy coding scheme (MEC) is proposed to achieve energy-efficiency in WNSNs. Unlike the existing minimum energy codes, MEC maintains the desired Hamming distance, while minimizing energy, in order to provide reliability. It is analytically shown that, with MEC, codewords can be decoded perfectly for large code distance, if source set cardinality, M is less than inverse of symbol error probability, 1/ps. Performance analysis shows that MEC outperforms popular codes such as Hamming, Reed-Solomon and Golay in average energy per codeword sense.Publication Open Access Leveraging frequency based salient spatial sound localization to improve 360 degrees video saliency prediction(Institute of Electrical and Electronics Engineers (IEEE), 2021) Çökelek, Mert; İmamoğlu, Nevrez; Özçınar, Çağrı; Department of Computer Engineering; Erdem, Aykut; Faculty Member; Department of Computer Engineering; College of Engineering; 20331Virtual and augmented reality (VR/AR) systems dramatically gained in popularity with various application areas such as gaming, social media, and communication. It is therefore a crucial task to have the knowhow to efficiently utilize, store or deliver 360° videos for end-users. Towards this aim, researchers have been developing deep neural network models for 360° multimedia processing and computer vision fields. In this line of work, an important research direction is to build models that can learn and predict the observers' attention on 360° videos to obtain so-called saliency maps computationally. Although there are a few saliency models proposed for this purpose, these models generally consider only visual cues in video frames by neglecting audio cues from sound sources. In this study, an unsupervised frequency-based saliency model is presented for predicting the strength and location of saliency in spatial audio. The prediction of salient audio cues is then used as audio bias on the video saliency predictions of state-of-the-art models. Our experiments yield promising results and show that integrating the proposed spatial audio bias into the existing video saliency models consistently improves their performance.Publication Open Access Uplink achievable rate maximization for reconfigurable intelligent surface aided millimeter wave systems with resolution-adaptive ADCs(Institute of Electrical and Electronics Engineers (IEEE), 2021) Xiu, Yue; Zhao, Jun; Di Renzo, Marco; Sun, Wei; Gui, Guan; Wei, Ning; Department of Electrical and Electronics Engineering; Başar, Ertuğrul; Faculty Member; Department of Electrical and Electronics Engineering; College of Engineering; 149116In this letter, we investigate the uplink of a reconfigurable intelligent surface (RIS)-aided millimeter-wave (mmWave) multi-user system. In the considered system, however, problems with hardware cost and power consumption arise when massive antenna arrays coupled with power-demanding analog-todigital converters (ADCs) are employed. To account for practical hardware complexity, we consider that the access point (AP) is equipped with resolution-adaptive analog-to-digital converters (RADCs). We maximize the achievable rate under hardware constraints by jointly optimizing the ADC quantization bits, the RIS phase shifts, and the beam selection matrix. The formulated problem is non-convex. To efficiently tackle this problem, a block coordinated descent (BCD)-based algorithm is proposed. Simulations demonstrate that an RIS can mitigate the hardware loss due to the use of RADCs, and that the proposed BCD-based algorithm outperforms state-of-the-art algorithms.Publication Open Access On the maximum coverage area of wireless networked control systems with maximum cost-efficiency under convergence constraint(Institute of Electrical and Electronics Engineers (IEEE), 2015) Kılınç, Deniz; Özger, Mustafa; Akan, Özgür Barış; PhD Student; College of EngineeringThe integration of wireless communication and control systems revealed wireless networked control systems (WNCSs). One fundamental problem in WNCSs is to have a wide coverage area. For the first time in the literature, we address this problem and we obtain the maximum coverage area by solving an optimization problem. In this technical note, we consider a WNCS where the output sensor measurements are transmitted over separate heterogeneous multi-hop wireless ad-hoc subnetworks. The observation process is divided into N parts and the system state is estimated using the Kalman filter. We present the critical arrival probability for a sensor measurement packet such that if the packet arrival probability is larger than the critical value, it is guaranteed that the estimator of the WNCS converges. We derive the maximum total coverage area of the heterogeneous wireless subnetworks having maximum cost-efficiency under the constraint of the convergence of the WNCS estimator.Publication Open Access Self-supervised monocular scene decomposition and depth estimation(IEEE Computer Society, 2021) Department of Computer Engineering; N/A; Güney, Fatma; Safadoust, Sadra; Department of Computer Engineering; Koç Üniversitesi İş Bankası Yapay Zeka Uygulama ve Araştırma Merkezi (KUIS AI)/ Koç University İş Bank Artificial Intelligence Center (KUIS AI); College of Engineering; Graduate School of Sciences and Engineering; 187939; N/ASelf-supervised monocular depth estimation approaches either ignore independently moving objects in the scene or need a separate segmentation step to identify them. We propose MonoDepthSeg to jointly estimate depth and segment moving objects from monocular video without using any ground-truth labels. We decompose the scene into a fixed number of components where each component corresponds to a region on the image with its own transformation matrix representing its motion. We estimate both the mask and the motion of each component efficiently with a shared encoder. We evaluate our method on three driving datasets and show that our model clearly improves depth estimation while decomposing the scene into separately moving components.Publication Open Access Finger-actuated microneedle array for sampling body fluids(Multidisciplinary Digital Publishing Institute (MDPI), 2021) Ahmadpour, Abdollah; Yetişen, Ali K.; Department of Mechanical Engineering; Taşoğlu, Savaş; Sarabi, Misagh Rezapour; Faculty Member; Department of Mechanical Engineering; KU Arçelik Research Center for Creative Industries (KUAR) / KU Arçelik Yaratıcı Endüstriler Uygulama ve Araştırma Merkezi (KUAR); Koç University Research Center for Translational Medicine (KUTTAM) / Koç Üniversitesi Translasyonel Tıp Araştırma Merkezi (KUTTAM); College of Engineering; Graduate School of Sciences and Engineering; 291971; N/AThe application of microneedles (MNs) for minimally invasive biological fluid sampling is rapidly emerging, offering a user-friendly approach with decreased insertion pain and less harm to the tissues compared to conventional needles. Here, a finger-powered microneedle array (MNA) integrated with a microfluidic chip was conceptualized to extract body fluid samples. Actuated by finger pressure, the microfluidic device enables an efficient approach for the user to collect their own body fluids in a simple and fast manner without the requirement for a healthcare worker. The processes for extracting human blood and interstitial fluid (ISF) from the body and the flow across the device, estimating the amount of the extracted fluid, were simulated. The design in this work can be utilized for the minimally invasive personalized medical equipment offering a simple usage procedure.Publication Open Access A queueing-theoretical delay analysis for intra-body nervous nanonetwork(Elsevier, 2015) Department of Electrical and Electronics Engineering; Abbasi, Naveed Ahmed; Akan, Özgür Barış; Faculty Member; Department of Electrical and Electronics Engineering; College of EngineeringNanonetworks is an emerging field of study where nanomachines communicate to work beyond their individual limited processing capabilities and perform complicated tasks. The human body is an example of a very large nanoscale communication network, where individual constituents communicate by means of molecular nanonetworks. Amongst the various intra-body networks, the nervous system forms the largest and the most complex network. In this paper, we introduce a queueing theory based delay analysis model for neuro-spike communication between two neurons. Using standard queueing model blocks such as servers, queues and fork-join networks, impulse reception and processing through the nervous system is modeled as arrival and service processes in queues. Simulations show that the response time characteristics of the model are comparable to those of the biological neurons.Publication Open Access Super-mode OFDM with index modulation(Institute of Electrical and Electronics Engineers (IEEE), 2020) Department of Electrical and Electronics Engineering; Doğukan, Ali Tuğberk; Başar, Ertuğrul; Faculty Member; Department of Electrical and Electronics Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 149116Orthogonal frequency division multiplexing (OFDM) with index modulation (OFDM-IM) appears as a promising multi-carrier waveform candidate for beyond 5G due to its attractive advantages such as operational flexibility and ease of implementation. However, OFDM-IM may not be a proper choice for 5G services such as enhanced mobile broadband (eMBB) since achieving high data rates is challenging because of its null subcarriers. One solution to enhance the spectral efficiency of OFDM-IM is the employment of multiple distinguishable constellations (modes) by also exploiting its null subcarriers for data transmission. This article proposes a novel IM technique called super-mode OFDM-IM (SuM-OFDM-IM), where mode activation patterns (MAPs) and subcarrier activation patterns (SAPs) are jointly selected and conventional data symbols are repetition coded over multiple subcarriers to achieve a diversity gain. For the proposed scheme, a low-complexity detector is designed, theoretical analyses are performed and a bit error rate (BER) upper bound is derived. The performance of the proposed system is also investigated through real-time experiments using a software-defined radio (SDR) based prototype. We show that SuM-OFDM-IM exhibits promising results in terms of spectral efficiency and error performance; thus, appears as a potential candidate for 5G and beyond communication systems.