56271 research outputs found
Sort by
Kız Kulesi ve Galata'nın aşkı
No full textUnutma İstanbul projesinin bir parçası olarak İstanbul Şehir Üniversitesi ile Türkiye Turing ve Otomobil Kurumu (TURİNG) işbirliği kapsamında Büşra Sürücü tarafından çekilmiştir.İstanbul Kalkınma Ajansı (TR10/16/YNY/0101) İstanbul Development Agency (TR10/16/YNY/0101
A model for an optical ring micro resonator towards nanoparticle detection
No full textOptical resonators have shown great potential as fundamental building blocks, in a wide range of photonic devices and applications including filters, optical multiplexers, logic gates, switches and sensors. A microresonator is defined by a micron-scale circular geometry (microdisk, microsphere, or microring) that confines light by total internal reflection. In the classical picture, optical resonances are coupled to the resonator when the length of the circular light path around the circumference matches an integer values of the wavelength. In a ring resonator, this resonance can be identified by a Lorentzian-shaped response in transmission spectra, typically obtained by delivering tunable laser light to the resonator via a tapered optical fiber positioned in the vicinity of the resonator. The resonance dips can be observed by measuring the transmission spectrum while sweeping the laser wavelength. As the wavelength of the laser matches the resonance wavelengths of the ring, light couples from the optical fiber to the microring, resulting in an intensity drop in Lorentzian line of transmission spectra monitored at the output of the waveguide. In this work, we have used an optical microring-based resonator, and developed a Multiphysics finite element model of this resonator for ultrasensitive biomolecule and nanoparticle detection. This model was developed using COMSOL Multiphysics v.5.2. We used an input tunable laser light at the input node of the waveguide at a wavelength of 1.55 μm and monitored at the transmitted light at the output node simulation. An illustration of the optical micro-ring resonator is given in Figure 1. Refractive index of the microring material was set to be at 1.5 for our simulations. During this process, we recorded the transmission spectra and located the exact position of the resonance wavelengths shifted towards higher wavelengths (red shift). Figure 2 illustrates the resonance wavelength value (transmission dip) by plotting the transmittance as a function of wavelength, for a ring radius of 83μm. We observed a linear correlation between increasing the radius of the microring and the red shift in the resonance wavelengths. In addition, we found that we can detect resonance shifts for very small changes in microring radius, as low as 10 nm. Figure 3 shows the resonance wavelength response with respect to the change in the microring radius. In summary, we have developed a model platform to simulate a microring resonator as a biosensor for biomolecule and nanoparticle detection based on the resonant frequency shift as a function of small microring size changes. In this respect, ring-based microresonator biosensors prospect to have a bright future in clinical diagnostics since they are label-free analytic detectors and also enabling the integration possibility on a chip-scale device
A wearable EEG-based serious game for focus improvement and diagnosing ADHD/ADD patients by EEG signals classification
No full textTezin basılısı İstanbul Şehir Üniversitesi Kütüphanesi'ndedir.Attention Deficit Hyperactivity Disorder (ADHD), characterized by the lack of attention and focus, is one of the most spread cognitive disorders. Since electroencephalogram (EEG)signalscarryextensiveinformationaboutcognitionskills,whichincludeattention, then the potential of using EEG signals for people with low attention span can be quite significant. EEG can be read using the new wireless EEG reading devices often used by Brain-computer Interface (BCI) researchers.
In parallel, serious games have been recently utilized for rehabilitating various cognitive and emotional deficits. In this thesis, we put the two things together, and we design a virtual reality serious game controlled using a wireless wearable EEG device to improve the attentiveness ability of people with ADHD/ADD. Our preliminary experiments with healthy subjects show an average improvement of 10% in engagement and 8% in focus for people using our EEG-controlled game compared to using the same game but keyboardcontrolled.
Furthermore, we investigate the integration of an EEG-controlled serious game that trains and strengthens patients’ attention ability while using machine learning to detect their attention level. The pilot experiments with healthy individuals show an accuracy of up to 96% in classifying the EEG data to detect the correct attention state during gameplay, and the extended experiments with ADHD patients show an accuracy up to 98% in classifying the patients EEG data.Abstract iv
Öz v
Acknowledgments vii
List of Figures x
List of Tables xi
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Problem Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.5 Research Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.6 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Background 6
2.1 EEG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 ADHD & ADD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 ASD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 GAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.5 SVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.6 KNN and DTW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.7 Cross-validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 Related Work 11
3.1 EEG-Based Therapies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 EEG-Based Serious games for ADHD . . . . . . . . . . . . . . . . . . . . . 13
3.3 Limitations of Related Work, and Our Contribution . . . . . . . . . . . . 14
4 Proposed Game: Design, Implementation, Preliminary Test 15
4.1 Game Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.1.1 Environment and the avatar . . . . . . . . . . . . . . . . . . . . . . 15
4.1.2 Initial Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.1.3 Training and Gameplay . . . . . . . . . . . . . . . . . . . . . . . . 18
4.1.4 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2 Experimentation and Results . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.1 Test Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.2 Preliminary Testing Results . . . . . . . . . . . . . . . . . . . . . . 21
5 Proposed Classification Models, Experiments, and Results 25
5.1 Data Wrangling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.1.1 Frequency Bands Data Filtering . . . . . . . . . . . . . . . . . . . 26
5.1.2 Data Windowing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.1.3 Feature Extraction (Extra Attributes) . . . . . . . . . . . . . . . . 26
5.1.4 Data Flattening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.2 Proposed Models: Emotiv-control VS. Keyboard-control Classification . . 27
5.2.1 Pre-recording Instructions . . . . . . . . . . . . . . . . . . . . . . . 28
5.2.2 Data Recordings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.2.3 Data Labeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.2.4 Model 1: Support Vector Machine (SVM) . . . . . . . . . . . . . . 29
5.2.4.1 Linear Kernel Model . . . . . . . . . . . . . . . . . . . . . 29
5.2.4.2 RBF Kernel Model . . . . . . . . . . . . . . . . . . . . . . 30
5.2.4.3 RBF Kernel Model with Extra Attributes . . . . . . . . . 30
5.2.4.4 One Subject Data Holdout . . . . . . . . . . . . . . . . . 31
5.2.4.5 Personalized Models: Model Per Subject . . . . . . . . . 31
5.2.4.6 Model for Attributes Importance . . . . . . . . . . . . . . 32
5.2.5 Model 2: Dynamic Time Warping (DTW) and K-Nearest Neighbours (KNN) Classifier . . . . . . . . . . . . . . . . . . . . . . . . . 32 5.2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5.3 Proposed Models: ADHD VS. non-ADHD Classification . . . . . . . . . . 33
5.3.1 Data Recordings and Data Labeling . . . . . . . . . . . . . . . . . 34
5.3.2 ADHD Classification Model: Support Vector Machine (SVM) . . . 35
5.3.2.1 RBF Kernel with Extra Attributes . . . . . . . . . . . . . 35
5.3.2.2 One ADHD Subject’s Data Holdout . . . . . . . . . . . . 36
5.3.2.3 Two Subjects’ Data Holdout (one ADHD and one nonADHD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6 Conclusion and Future Work 40
A Ethics Committee Decision 42
Bibliography 4
Turkish challenges for low-carbon society:current status, government policies and social acceptance
No full textTurkey’s rapid economic growth followed by increasing gap between domestic energy production and
consumption signifies the large-scale integration of low-carbon and renewable resources into energy system.
Turkey’s energy demand is increasing and supplying fossil energies such as oil and gas is becoming more
challenging due to the lack of domestic hydrocarbon resources. The country is facing two main issues; first,
maintaining energy balance and second, handling environmental issues. Nevertheless, Turkey has substantial
renewable and low-carbon energy resources with co-benefits for the country’s fast growing economy and
environmental sustainability. This paper aims to present a comprehensive update on Turkey’s renewable and
low-carbon energy sector. It reviews not only the current status of the technology, policy and economy, but also
provides an insight into the social aspects as well. From a policy point of view, the paper critically discusses
Turkey’s energy strategy and perspectives and from the social point of view, the paper argues that there is a
research gap in recent literature regarding public awareness and social acceptance of low-carbon energy
technologies. Also, the future challenges for promoting low-carbon society in Turkey are discussed in terms of
sustainable development and socio-environmental conditions
Detection of attack-targeted scans from the apache HTTP server access logs
No full textTezin basılısı İstanbul Şehir Üniversitesi Kütüphanesi'ndedir.A web application could be visited for different purposes. It is possible for the resources (web pages, images, flash objects, etc.) provided by a web application to be visited by a regular user as a normal (natural) visit, to be viewed by crawlers, bots, spiders etc. for indexing purposes, or to be exploratory scanned by malicious users prior to an attack. In this thesis, a method; that can be used to detect attack-oriented scans and to distinguish them from two other types of visits, has been proposed. This method is also implemented with a Python script.
To detect these malicious scans, there are different detection mechanisms such as web application firewalls, machine learning based predictive detection techniques etc. In the scope of the thesis, static access log files of Apache (or ISS) web servers are used as an input and attack situations have been tried to be determined in such a way as to allow the examination of the past data. An attack targeted website scan can be viewed as a phase of a potential attack and can lead to more attack detection as compared to traditional detection methods. In addition to web scan detections, an extra rule set has been inserted to detect two largely common injection attacks in OWASP Top 10 Vulnerabilities.
Detection rules proposed in the thesis have been applied on sample data sets and evaluations have been made on these data sets that are both closed and open to external traffic. Moreover, results have been analysed in terms of accuracy rate, precision etc. to compare presented method and other commonly used detection techniques, and to remark its disadvantages and advantages. Furthermore, various tests have been made on log samples from real systems. Lastly, several suggestions about further development have also been discussed.Declaration of Authorship ii
Abstract iv
Öz v
Acknowledgments vii
List of Figures x
List of Tables xi
Abbreviations xii
1 Introduction 1
1.1 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Basic Concepts 7 2.1 Apache HTTP Server Access Log . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 Access Log Formats . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Hypertext Transfer Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 HTTP Header Fields . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 HTTP Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 HTTP Status Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.5 Uniform Resource Identifiers (URI) . . . . . . . . . . . . . . . . . . . . . . 13
2.6 Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6.1 UTF-8 Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.6.2 Hex/Base 16 Encoding . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7 Robots Exclusion Standard . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 System Model 17
3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2 Data and Log Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2.1 Virtual Machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.2 Guest Operating Systems . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.3 Web Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.3.1 Apache HTTP Server . . . . . . . . . . . . . . . . . . . . 18
3.2.3.2 Apache Tomcat . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.4 Damn Vulnerable Web Application (DVWA) . . . . . . . . . . . . 19
3.2.5 Web Vulnerability Scanners . . . . . . . . . . . . . . . . . . . . . . 20
3.2.5.1 Acunetix . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.5.2 Netsparker . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.5.3 Web Application Attack and Audit Framework (W3AF) . 20
3.3 Rules and Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4 Results 26
4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.2 Model Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.3 Scan Detection on Live Data . . . . . . . . . . . . . . . . . . . . . . . . . 28
5 Conclusions 33
Bibliography 3
Exploring the suitability of electric vehicles in the United States
No full textThis study explores suitability of battery electric vehicles in the United States by considering their potential market share and operations costs as well as the state-specific variations in electricity generation profiles, given current government policies and the social acceptability of the technology. A performance assessment is developed to compare each state and identify major policy efforts that are needed to increase the environmental and economic competitiveness of electric vehicles. A novel multi-criteria decision-support framework, integrating Life Cycle Assessment, Data Envelopment Analysis, and Agent Based Modeling, is developed. To this end, the environmental and economic impacts of battery electric vehicles are calculated based on three scenarios: an average electricity generation mix, a marginal electricity generation mix, and a solely renewable energy mix with 100% solar. The states are classified, each requiring different policy strategies, in accordance with their performance scores. The results provide important insights for advancing transportation policies and a novel framework for multi-criteria decision-making in the future analyses
EMDR therapy on trauma-based restrictive eating cases
No full textEating Disorders (ED) affect an individual's physical and mental health with abnormal eating habits. Traumatic life events may underlie the development of ED as many studies document [1]. In the present study, we examined the effectiveness of EMDR therapy that was originally used to treat Post-Traumatic Stress Disorder (PTSD) [2], on restrictive eating symptoms associated with trauma. Cases B.Ö. (18) and B.S. (20) came to the clinic consecutively for resembling complaints about the sense of food sticking in their throat, breathing difficulties, raised heart beatings, unease to swallow, and choking fear. The patients revealed past traumatic events about being out of breath while swallowing their bites. Then, they have started to restrict their food intake and the types of food consumed, which led to emaciation with health problems, interrupted daily routines, and social isolation; meeting the diagnosis of avoidant/restrictive food intake disorder (ARFID) in DSM-5. Due to traumatic experiences, EMDR therapy was applied
An intelligent cloud-based data processing broker for mobile e-health multimedia applications
No full textMobile e-health applications provide users and healthcare practitioners with an insightful way to check users/patients’ status and monitor their daily calorie intake. Mobile e-health applications provide users and healthcare practitioners with an insightful way to check users/patients’ status and monitor their daily activities. This paper proposes a cloud-based mobile e-health calorie system that can classify food objects in the plate and further compute the overall calorie of each food object with high accuracy. The novelty in our system is that we are not only offloading heavy computational functions of the system to the cloud, but also employing an intelligent cloud-broker mechanism to strategically and efficiently utilize cloud instances to provide accurate and improved time response results. The broker system uses a dynamic cloud allocation mechanism that takes decisions on allocating and de-allocating cloud instances in real-time for ensuring the average response time stays within a predefined threshold. In this paper, we further demonstrate various scenarios to explain the workflow of the cloud components including: segmentation, deep learning, indexing food images, decision making algorithms, calorie computation, scheduling management as part of the proposed cloud broker model. The implementation results of our system showed that the proposed cloud broker results in a 45% gain in the overall time taken to process the images in the cloud. With the use of dynamic cloud allocation mechanism, we were able to reduce the average time consumption by 77.21% when 60 images were processed in parallel
Redundancy allocation based on the weighted mismatch-rate slope for multiple description video coding
No full textMultipledescriptioncoding(MDC)isarobustcoding technique for video transmission over error prone networks, whereby the video is encoded into multiple descriptions with some redundancy between the descriptions. This redundancy leads to error resiliency in the case of packet loss during the network transport. However, the amount of this redundancy has a critical role in MDC performance. Therefore, a crucial problem in MDC is to find what the optimum amount of redundancy budget is, and then how this redundancy budget can be optimally allocated to the frames. To solve this problem, we propose a scheme in which the redundancy budget is allocated to the frames based on the weighted mismatch-rate slopes so that this additional bitrate can attain maximum distortion reduction. The redundancy is added gradually so that fine tuning of the utilized bitrate is achievable. We have verified our proposed scheme by implementing it in H.264/AVC reference software JM16.0, and running experiments against two representative reference methods. Our experiments showthatourschemenotonlyminimizestheend-to-enddistortion witharate-distortionperformancethatisbetterthanthereference methods, especially for high PLRs, but also entirely uses the available bandwidth, unlike the reference methods