TOBB ETU GCRIS Database
Not a member yet
9778 research outputs found
Sort by
Obstrüktif Uyku Apnesinin Derin Ögrenme Kullanilarak Tahmin Edilmesi
No full textSleep apnea is defined as the cessation of breathing for at least ten seconds and known as a common sleep disorder. Obstructive sleep apnea (OSA) is the most common type of sleep apnea that occurs due to obstruction in the airway. Besides the detection of sleep apnea with the help of developed algorithms, early prediction of this syndrome is important in order to prevent serious health problems and life-threatening situations. With the prediction models working with high accuracy, it will be possible to prevent the possible risks without experiencing them by stimulating the OSA patients before the syndrome occurs and waking them up from sleep. In this thesis, models that predict apnea using electrocardiographic signals of patients diagnosed with obstructive slep apnea by using Convolutional Neural Networks (CNN) are presented. The first model is the training from scratch with pre-trained architectures. The second model is the study of using the transfer learning method with pre-trained architectures. The third model is the study in which the new results observed with the changes made on the architecture that performed the best prediction performance in the first two models. In the last model, the results are presented when the features obtained from the deep learning architecture proposed in the third model are classified by Support Vector Machines, Random Subspace k-Nearest Neighborhood and Random Subspace Discriminant Analysis methods instead of the architecture's own classifier. The high accuracy findings observed at the end of the study show that the proposed model can be used as a good indicator for the prediction of OSA syndrome.Nefes alışverişinin en az on saniye boyunca durması olarak tanımlanan uyku apnesi, günümüzde sık karşılaşılan bir uyku hastalığı olarak bilinmektedir. Obstrüktif uyku apnesi (OUA), solunum yolunda tıkanmaya bağlı gerçekleşen en yaygın uyku hastalıklarından biridir. Bu sendromun geliştirilen modeller yardımıyla otomatik olarak tespit edilebilmesinin yanında ön görülebilmesi de ciddi seviyelerde sağlık problemlerini ve hayati tehlikeyle karşı karşıya kalma durumunu önlemek açısından önemlidir. Yüksek doğrulukla çalışan tahmin modellerinin geliştirilmesi ile OUA rahatsızlığı yaşayan kişilerin sendrom anı gelmeden uyarılması ve uykudan uyandırılması ile olası risklerin yaşanmadan önlenmesi mümkün olabilecektir. Bu tez çalışmasında, derin öğrenme yöntemlerinden biri olan Evrişimsel Sinir Ağları (Convolutional Neural Networks, CNN) ile OUA tanısı konmuş hastalara ait elektrokardiyografi sinyalleri kullanılarak apne tahmini yapan modeller sunulmuştur. Bu modellerden birincisi, önceden eğitilmiş mimariler ile yapılan sıfırdan eğitim çalışmasıdır. İkinci model, önceden eğitilmiş mimariler ile transfer öğrenme yöntemi kullanılarak yapılan çalışmadır. Üçüncü model, ilk iki modelde gözlemlenen bulgulara bağlı olarak önerilmiş yeni bir modeldir. Son modelde ise üçüncü modelde önerilen derin öğrenme mimarisinden elde edilen özniteliklerin, mimarinin kendi sınıflandırıcısı yerine Destek Vektör Makineleri, Rastgele Alt Uzay k-En yakın Komşuluk ve Rastgele Alt Uzay Diskriminant Analizi yöntemleri ile sınıflandırıldığı durumda gözlemlenen sonuçlar sunulmuştur. Çalışmanın sonucunda gözlemlenen yüksek doğruluktaki bulgular, önerilen modellerin OUA tahmininde iyi bir belirteç olarak kullanılabileceğini göstermektedir
The Contagion Dynamics of Vaccine Skepticism
No full textIn this manuscript, we discuss the spread of vaccine refusal through a non-linear mathematical model involving the interaction of vaccine believers, vaccine deniers, and the media sources. Furthermore, we hypothesize that the media coverage of disease-related deaths has the potential to increase the number of people who believe in vaccines. We analyze the dynamics of the mathematical model, determine the equilibria and investigate their stability. Our theoretical approach is dedicated to emphasizing the importance of convincing people to believe in the vaccine without getting into any medical arguments. For this purpose, we present numerical simulations that support the obtained analytical results for different scenarios. © 2022, Hacettepe University. All rights reserved
Neurological Effects of Sars-Cov and Neurotoxicity of Antiviral Drugs Against Covid-19
No full textSevere Acute Respiratory Syndrome (SARS) is caused by different SARS viruses. In 2020, novel coronavirus (SARS-CoV-2) led to an ongoing pandemic, known as Coronavirus Disease 2019 (COVID-19). The disease can spread among individuals through direct (via saliva, respiratory secretions, or secretion droplets) or indirect (through contaminated objects or surfaces) contact. The pandemic has spread rapidly from Asia to Europe and later to America. It continues to affect all parts of the world at an increasing rate. There have been over 92 million confirmed cases of COVID-19 by mid-January 2021. The similarity of homological sequences between SARS-CoV-2 and other SARSCoVs is high. In addition, clinical symptoms of SARS-CoV-2 and other SARS viruses show similarities. However, some COVID-19 cases show neurologic signs like headache, loss of smell, hiccups and encephalopathy. The drugs used in the palliative treatment of the disease also have some neurotoxic effects. Currently, there are approved vaccines for COVID-19. However, there is a need for specific therapeutics against COVID-19. This review will describe the neurological effects of SARS-CoV-2 and the neurotoxicity of COVID-19 drugs used in clinics. Drugs used in the treatment of COVID-19 will be evaluated by their mechanism of action and their toxicological effects
On the Existence of Equivalent-Input and Multiple Integral Augmentation Via H-Infinity Synthesis for Unmatched Systems
No full textIn this paper, the existence of a solution for the transformation of the disturbances from the unmatched cases to the matched one is investigated. The usage of matched/unmatched disturbance notions and the underlying assumptions are clarified. Then, a simplified definition is introduced to obtain a set of performance metrics to be used in observer design. Using bilinear pole shifting and multiple integral augmentation to the plant, not only the stabilizability/detectability conditions but also infinity-norm bounds for unstable MIMO systems are derived. Then, the solvability of the augmented Hamiltonian matrices to get stabilizing solutions via standard H?-Synthesis is explained. Finally, the solutions, definitions, and assumptions are validated through numerical examples. © 2022 IS
Machine Learning for Failure Analysis: a Mathematical Modelling Perspective
No full textFailure analysis is an interdisciplinary and exciting research area that lies at the interface between mathematics, physics, and materials science. In this context, from the mathematical perspective, failure is a gradual or sudden loss of the ability to operate. Failure analysis leads to many interesting mathematical challenges spanning from model development to mathematical foundations including simulations and machine learning approaches. From a mathematical modelling viewpoint, there exists a variety of approaches that involve predicting whether failure will occur, computing the time until failure takes place, classifying failure modes, detecting anomalous behaviour, assessing the extent of deviation or degradation from an expected normal operating condition, among others. The data era has brought about approaches that employ machine learning (ML) methods to not only detect failure but also make predictions about the reliability (i.e., the ability to function without failure) of devices. They constitute a wide spectrum of models, ranging from survival models to investigate components ageing to Bayesian networks for anomaly detection, including generative neural networks to detect for example the degradation of a material. In this chapter, we give a description and illustration of the assumptions and basic models of ML and present a range of applications. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG
Kompozit Malzeme Üretiminde Kullanilan Paralel Firinlarin Çizelgelenmesi için Bir Optimizasyon Modeli ve Sezgisel Çözüm Yaklasimi Gelistirilmesi
No full textWe tackle a scheduling problem encountered in a real production department that produces composite parts in an aircraft manufacturing plant. Two main steps in the production process of composite parts are mounting the composite parts on molds and then heat treatment of the parts in pressurized parallel ovens, called autoclaves. Parts have different process requirements in terms of heating level, pressure, and time. Only the compatible parts can go into the same autoclaves together in a batch. The scheduling problem is about the second step of the process and it involves batching the parts together and then scheduling batches into the autoclaves. The problem has several different types of constraints, such as the capacity of autoclave in terms of space and thermocouple, the number of molds available for the process, due dates, the earliest and latest processing time for parts, and the sequence status of parts. The objective is taken as the minimization of the energy consumption since the autoclaves consume high levels of electricity. Closest problem to our problem in the literature is called batch scheduling with incompatible job families. In this study we introduced a mathematical model of the problem and preliminary solutions were obtained under different scenarios. Since the problem is in the NP-hard category, solutions cannot be obtained in a reasonable time for complex problems. For this reason, the K-means algorithm is developed for the problem, which divides the works into batches then schedules the batches to the furnaces, obtains the first feasible solution, and improves the schedules by the variable neighbor search (DKA) algorithm. In this way parameter sensitivity analysis of the mathematical model and the performance of the developed heuristic tested. As a result of the tests, it was observed that the heuristic algorithm deviated from the optimal by 5,12732% on average.Bu çalisma kapsaminda havacilik ve uzay sanayine yönelik çalisan ve kompozit parçalar üreten gerçek bir üretim departmaninda karsilasilan bir çizelgeleme problemi ele alinmaktadir. Kompozit parçalarin üretim sürecindeki iki ana adim, kompozit parçalarin kaliplara montesi ve ardindan parçalarin kaliplar içerisinde otoklav adi verilen basinçli paralel firinlarda isil islem görmesidir. Parçalar, isi seviyesi, basinç ve süre açisindan farkli islem gereksinimlerine sahiptirler. Yalnizca bu özelliklere göre uyumlu parçalar bir arada ayni partiye girebilir. Çizelgeleme problemi, sürecin ikinci adimi ile ilgilidir ve parçalarin birlikte gruplandirilip partilerin olusturulmasini ve ardindan otoklav adli firinlara giren partilerin firinlarda çizelgelenmesini içerir. Problemin otoklavlarin alan ve termocouple kapasiteleri, süreçte kullanilan kalip sayisi, parçalarin teslim tarihi, en erken ve en geç isleme alinabilecekleri zaman, ardisiklik durumu gibi pek çok kisitlari vardir. Otoklavlar yüksek düzeyde elektrik tükettigi için problemin amaci kullanilan parti sayisini en azlayarak enerji tüketiminin en aza indirilmesidir. Problem literatürde uyumsuz is aileleri ile parti çizelgeleme olarak geçmektedir. Tez kapsaminda problemin matematiksel modeli gelistirilmis ve farkli senaryolar altinda ön çözümler elde edilmistir. Problem NP-zor kategoride oldugundan yüksek boyutlu problemler için makul sürede çözüm elde edilememektedir. Bu sebeple problem için K-ortalama algoritmasi ile isleri partilere bölen, sonra partileri firinlara çizelgeleyip ilk olurlu çözümü elde eden ve degisken komsu arama (DKA) algoritmasi ile elde edilen çizelgeleri iyilestiren bir sezgisel algoritma gelistirilmistir. Problem farkli senaryolarda denenerek olusturulan matematiksel modelin parametre hassasiyet analizi ve gelistirilen sezgiselin performansi test edilmistir. Yapilan testler sonucu sezgisel algoritmanin ortalamada optimalden %5,12732 saptigi gözlemlenmistir
Paylaşımlı Çalışma Mekânlarında Kişiselleştirme Problemine Yeni Bir Öneri: Genişletilmiş Gerçeklik Uygulamalarının Kullanımı
No full textBu araştırmada, paylaşımlı çalışma mekânlarının kişiselleştirilememesi problemi ve kullanıcıların dijital bir kişiselleştirme yöntemi ile anonim mekanları kendilerine ait birer çalışma mekanına dönüştürebilme potansiyelleri incelenmiştir. Yapılan çalışmalar ile bu probleme günümüz teknolojik imkânları ve değişen çalışma kültürünün gereklilikleri doğrultusunda genişletilmiş gerçeklik ile yeni bir deneyim önerisi sunmak amaçlanmıştır. Nitel araştırma yöntemlerinden durum çalışması deseni ile gerçekleştirilen bu çalışmada veri toplama aracı olarak gözlem ve görüşmelerden yararlanılmıştır. Araştırma bulgularına göre kullanıcıların uygulama sayesinde, uygulama süresince hissedilen seçim ve yerleşim yapma özgürlüğü, özel alan kurgusu yaratabilme imkânı sağlaması gibi avantajlar yarattığı; bu nedenle motivasyon ve aidiyet durumlarını olumlu yönde etkilediği sonucu ortaya çıkmıştır. Kullanıcıların fiziksel mekânın hibrit deneyimdeki rolü üzerinden yaptıkları değerlendirmeler sonucunda; fiziksel mekânın yaratılan dijital kurgu için mekânsal referanslar sağladığı, bu nedenle hibrit mekânın fiziksel bağlamından tamamen kopmasını engellediği sonucu ortaya çıkmıştır. Elde edilen sonuçlar doğrultusunda araştırmanın; genişletilmiş gerçeklik teknolojisinin bu alanda özgün bir kişiselleştirme aracı olabilme potansiyeli konusunda daha sonraki araştırmalar için bir veri kaynağı olması hedeflenmiştir.In this research, the problem of non-personalization of shared workspaces and the potential of users to transform anonymous spaces into their own workspaces with a digital personalization method are examined. With the studies conducted, it is aimed to offer a new experience proposal to this problem with augmented reality in line with today's technological possibilities and the requirements of the changing work culture. In this study, which was carried out with a case study design from qualitative research methods, observations and interviews were used as data collection tools. According to the findings of the research, it was concluded that the application created advantages such as the freedom of choice and placement felt during the application, the opportunity to create a private space fiction; therefore, it positively affected the motivation and belonging status of the users. As a result of the users' evaluations on the role of the physical space in the hybrid experience; it was concluded that the physical space provides spatial references for the digital fiction created, thus preventing the hybrid space from being completely detached from its physical context. In line with the results obtained, the research is aimed to be a data source for further research on the potential of augmented reality technology as a unique personalization tool in this field
Hydrogen Electrical Vehicles
No full textHydrogen electrical vehicles are an essential component of the “Green New Deal” and this book covers cutting-edge technologies designed for fuel-cell-powered cars. The realization of the decision of 28 countries to keep global warming at 2 degrees and below, which is stated in the Paris Agreement, and the achievement of minimizing CO2 emissions, can only be accomplished by establishing a hydrogen ecosystem. A new geopolitical order is envisaged, in which sectors dealing with energy production, distribution, and storage, thus decreasing the carbon footprint, are reconstructed. In short, an economic order with new tax regulations is being created in which the carbon footprint will be followed. This global effort called the “Green Deal” is defined as a new growth strategy aiming at net-zero CO2 emissions. We know that the total share of transportation in CO2 emissions is about 24%. Therefore, efforts for reducing emissions must include utilizing hydrogen in transport. The subjects covered in the book include: An introduction to hydrogen and electrical vehicles; Hydrogen storage and compression systems; Hydrogen propulsion systems for UAVs; Test and evaluation of hydrogen fuel cell vehicles; Hydrogen production and PEM fuel cells for electrical vehicles; The power and durability issues of fuel cell vehicles. Audience The book will attract readers from diverse fields such as chemistry, physics, materials science, engineering, mechanical and chemical engineering, as well as energy-focused engineering and hydrogen generation industry programs that will take advantage of using this comprehensive review of the hydrogen electrical vehicles. © 2023 Scrivener Publishing LLC
Davadan Feragat, Ölen Eşin Mirasçılarının Tmk M. 181/2 Hükmüne Göre Davayı Devam Ettirmesine Engel Olabilir Mi?
No full textBoşanma davası devam ederken ölen davalı eşin mirasçılarının, TMK m. 181/2 hükmü uyarınca davaya devam etmesi halinde, sağ kalan davacı eş, davadan feragat etmek suretiyle, ölenin mirasçıların bu haklarını kullanmalarına engel olamaz. Bu halde davadan feragate engel açık bir hüküm bulunmasa ve teorik olarak konusuz kalan bir davadan feragat mümkün olsa bile, söz konusu feragat beyanı, hem belirtilen normun amacına hem de dürüstlük kuralı ve hakkın kötüye kullanılması yasağına aykırı olacağı için hukuken sonuç doğuramaz