56271 research outputs found
Sort by
Feature selection via binary simultaneous perturbation stochastic approximation
No full textFeature selection (FS) has become an indispensable task in dealing with today’s highly complex pattern recognition problems with massive number of features. In this study, we propose a new wrapper ap-proach for FS based on binary simultaneous perturbation stochastic approximation (BSPSA). This pseudogra dientdescent stochastic algorithm starts with an initial feature vector and moves toward the optimal feature vector via successive
iterations. In each iteration, the current feature vector’s individual com- ponents are perturbed simultaneously by offsets from a qualified probability distribution. We present computational experiments on datasets with numbers of features ranging from a few dozens to thousands using three widely-used classifiers as wrappers: nearest neighbor, decision tree, and linear support vectormachine. We
compare our methodology against the fullset of features as well as a binary genetic algorithm and sequential FS methods using cross-validated classification error rate and AUC as theperformance criteria. Our results indicate that features selected by BSPSA compare favorably to alter- native methods in general and BSPSA can yield superior feature sets for datasets with tens of thousands of features by examining an extremely small fraction of the solution space. We are not aware of any other wrapper FS methods that are computationally feasible with good convergence properties for such large datasets
Energy-climate-manufacturing nexus:new insights from the regional and global supply chains of manufacturing industries
No full textThe main objectives of this research are to improve our understanding of energy-climate-manufacturing nexus within the context of regional and global manufacturing supply chains as well as show the significance of full coverage of entire supply chain tiers in order to prevent significant underestimations, which might lead to invalid policy conclusions. With this motivation, a multi region input–output (MRIO) sustainability assessment model is developed by using the World Input–Output Database, which is a dynamic MRIO framework on the world’s 40 largest economies covering 1440 economic sectors. The method presented in this study is the first environmentally-extended MRIO model that harmonizes energy and carbon footprint accounts for Turkish manufacturing sectors and a global trade-linked carbon and energy footprint analysis of Turkish manufacturing sectors is performed as a case study. The results are presented by distinguishing the contributions of five common supply chain phases such as upstream suppliers, onsite manufacturing, transportation, wholesale, and retail trade. The findings showed that onsite and upstream supply chains are found to have over 90% of total energy use and carbon footprint for all industrial sectors. Electricity, Gas and Water Supply sector is usually found to be as the main contributor to global climate change, and Coke, Refined Petroleum, and Nuclear Fuel sector is the main driver of energy use in upstream supply chains. Overall, the largest portion of total carbon emissions of Turkish manufacturing industries is found in Turkey’s regional boundary that ranged between 40% and 60% of total carbon emissions. In 2009, China, United States, and Rest-of-the-World’s contribution is found to be more than 50% of total energy use of Turkish manufacturing. The authors envision that a global MRIO framework can provide a vital guidance for policy makers to analyze the role of global manufacturing supply chains and prevent significant underestimations due to inclusion of limited number of tiers for sustainable supply chain management research
Secure and efficient biometric authentication based on advanced cryptographic primitives
No full textTezin basılısı İstanbul Şehir Üniversitesi Kütüphanesi'ndedir.In this thesis, we study secure biometric authentication protocols and mitigation methods
against password database breaches. Biometric identifiers such as fingerprint of a user
is inherently unique and can be used to authenticate individuals. However, biometric
identifiers cannot be replaced with new ones, once they are compromised. Therefore,
biometric authentication systems that use biometric data in plain form raises security
and privacy issues. In this thesis, we give a survey of state of the art protocols designed
for secure biometric authentication. Hamming distance is generally used to compare
biometric feature veectors. As a practical example, we analyze a cryptographic protocol
of Bringer et al. (called SHADE) which aims to securely compute a Hamming distance
computation based on Committed Oblivious Transfer protocol. We show that SHADE is
in fact insecure in the malicious model. We mount different attacks to the protocol and
introduce mitigation techniques that makes the protocol indeed secure. Furthermore, we
also analyze the protocol from the efficiency perspective and show that the complexity
of the protocol can be significantly improved.
Another important security problem in authentication is the password database breach.
In the last few years, several password databases were breached and millions of user
names and password hashes were released. With the recent technological advancements
like using graphical-processing unit (GPU) in computation, it is comparably easier to
invert password hashes. Once the password database has been breached and passwords
have been recovered, no server can detect any illegitimate user authentication. In order
to thwart these kinds of threats, Juels and Rivest developed the Honeywords system.
In this system, each user is associated with multiple possible passwords but only one of
them is genuine. This thesis will analyze the security of the Honeywords system as well
as from functionality perspective. The authors point out that the Honeyword system
cannot withstand active attacks, i.e., code modification of the system. We introduce an
enhanced model which solves this open problem. We propose some improvements for
determining the number of Honeywords per user, generating typo-safe honeywords and
managing old passwords. Finally, the security and efficieny analysis has been discussed.Abstract iii
Öz iv
Acknowledgments vi
List of Figures ix
List of Tables x
Abbreviations xi
1 Introduction 1 1.1 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 List of Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Preliminaries 9
2.1 Mathematical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.1 COT versus VOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3 Password Related Attacks . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4 Security and Privacy Model . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3 Privacy Issue of the SHADE Protocol and Efficiency Improvements 16
3.1 The Basic and the Full Scheme of SHADE . . . . . . . . . . . . . . . . . . 16
3.1.1 The Basic Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.2 The Full Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2 Security and Efficiency Analysis of SHADE . . . . . . . . . . . . . . . . . 18
3.2.1 Attack to the Full Scheme . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.2 A Special Case: Apply the Attack to Biometric Authentication Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2.3 Apply the Generic for Uniformly Distributed Inputs . . . . . . . . 21
3.2.4 Our Solution for the Attack . . . . . . . . . . . . . . . . . . . . . . 22
3.2.4.1 More Efficient Solution for Biometric Authentication . . . 23
3.2.5 Efficiency Enhancements . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2.5.1 Efficiency Improvement Using VOT . . . . . . . . . . . . 23
3.3 Our Fixed and Improved Scheme . . . . . . . . . . . . . . . . . . . . . . . 24
3.4 Security Analysis of Our Scheme . . . . . . . . . . . . . . . . . . . . . . . 25
3.5 Complexity Analysis Of Our Fixed Protocol . . . . . . . . . . . . . . . . . 27
4 Enhanced Honeywords System 29
4.1 Honeywords System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.2 Improvements for Honeywords System . . . . . . . . . . . . . . . . . . . . 31
4.2.1 Number of Honeywords . . . . . . . . . . . . . . . . . . . . . . . . 31
4.2.2 Typo-Safe Honeyword Generation . . . . . . . . . . . . . . . . . . 32
4.2.3 Old Passwords Problem . . . . . . . . . . . . . . . . . . . . . . . . 34
4.3 Solution to an Open Problem: Active Attacks Against Honeywords System 34
4.3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.2 Adversarial Capabilities . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.3 The Proposal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3.4 Security Analysis of the New Proposed Model . . . . . . . . . . . . 37
5 Conclusion 39
Bibliography 4
Enhanced dissolution of liquid microdroplets in the extensional creeping flow of a hydrodynamic trap
No full textA novel noncontact technique based on hydrodynamic trapping is presented to study the dissolution of freely suspended liquid microdroplets into a second immiscible phase in a simple extensional creeping flow. Benzyl benzoate (BB) and n-decanol microdroplets are individually trapped at the stagnation point of a planar extensional flow, and dissolution of single microdroplets into an aqueous solution containing surfactant is characterized at different flow rates. The experimental dissolution curves are compared to two models: (i) the Epstein–Plesset (EP) model which considers only diffusive mass transfer, and (ii) the Zhang–Yang–Mao (ZYM) model which considers both diffusive and convective mass transfer in the presence of extensional creeping flow. The EP model significantly underpredicts the experimentally determined dissolution rates for all experiments. In contrast, very good agreement is observed between the experimental dissolution curves and the ZYM model when the saturation concentration of the microdroplet liquid (cs) is used as the only fitting parameter. Experiments with BB microdroplets at low surfactant concentration (10 μM) reveal cs values very similar to that reported in the literature. In contrast, experiments with BB and n-decanol microdroplets at 10 mM surfactant concentration, higher than the critical micelle concentration (CMC) of 5 mM, show further enhancements in microdroplet dissolution rates due to micellar solubilization. The presented method accurately tests the dissolution of single microdroplets into a second immiscible phase in extensional creeping flow and has potential for applications such as separation processes, food dispersion, and drug development/design
Personalized psychotherapy for adult depression:a meta-analytic review
No full textPersonalized medicine is aimed at identifying which characteristics of an individual predict the outcome of a specific treatment, in order to get a better match between the individual and the treatment received. We conducted a systematic review and meta-analysis of randomized trials comparing two psychotherapies directly in a group of depressed patients with a specific characteristic. We focused on the six most examined types of psychotherapy for adult depression. Our searches resulted in 41 studies with 2,741 patients who met inclusion criteria. These 41 studies examined 27 specific characteristics of patients. Power calculations indicated that we would need 4 studies for each characteristic to find a clinically relevant effect size set at g = 0.50 and 16 studies for an effect size of 0.24. Only 3 patient characteristics were found to have sufficient power and to significantly moderate treatment outcomes. Cognitive-behavioral therapy was found to be more effective than other therapies in older adults (g = 0.29), in patients with comorbid addictive disorders (g = 0.31), and in university students (g = 0.46). Risk of bias was considerable in most of the included studies. It was estimated that it will take another 326 years to have sufficient statistical power for showing an effect size of g = 0.50 of the 27 characteristics, and 1,372 years to show an effect size of 0.24. Although several dozens of studies have compared the effects of psychotherapies in specific target groups, we will need to develop more powerful alternatives to comparative outcome studies in order to identify personalized treatments for depression
Settlement of Geyikli Baba in İnegöl, his relationship with Orhan Ghazi and refutation of the arguments on his being a heterodox in the light of direct evidences
No full textBildiri, yazarın izniyle açık erişim kullanıma açılmıştır.Osmanlı Uç Sultanlığı’nın ilk ve en önemli yerleşim merkezlerinden biri olan İnegöl ucuna yerleşerek, Bursa muhâsarası sırasında dervişleriyle birlikte Kızılkilise (Kemaliye/Keles) taraflarını fetheden Geyikli Baba ile Orhan Gâzî arasında yaşanan ilginç bir vak‘a, XV. yüzyıl başlarında düzenlenmiş olan eski bir belgeye yansımıştır. Üç parça hâlinde günümüze intikâl eden bu belgenin yalnız ilk parçasında yer alan kayıtlara dayanılarak, XX. yüzyıl başlarında ilkin Geyikli Baba’nın rakı ve şarap içen heterodoks bir derviş olduğuiddiâsı ortaya atılmış; daha sonra ise farklı tarih araştırmacıları tarafından bu isâbetsiz yaklaşım, Baba’nın çağdaşı diğer Türkmen babalarının da karakteristik bir özelliği imiş noktasına taşınmıştır. Bu bildiride,Geyikli Baba ve XIV. yüzyıl Türkmen babalarının heterodoks bir yaşam tarzını benimsedikleri iddiâsının temelini teşkil eden bu tartışmalı belgeden, bu belgede anlatılanları tekrâr eden anonim bir yazmadaki ortak bilgilerden ve dönemin diğer tarihî materyallerinden hareketle, İnegöl ucuna yerleşen Geyikli Baba’nın Bursa fethi sırasındaki ve sonrasındaki faaliyetleri, Orhan Gâzî ile yakın ilişkisi ve burada kendisine vakıflar tahsis edilmesi konuları mercek altına alınarak, Baba’nın heterodoksluğuna yönelik mesnedsiz iddiâlar kesin deliller ışığında çürütülmeye çalışılacaktır
A highly integrated wideband LNA with multiple inputs for multi-band mobile devices
No full textAbstract—This paper presents a wideband common gate LNA (CG-LNA) that provides a competitive and highly integrated front-end solution for mobile handset devices. The proposed design supplies multiple inputs in a single LNA design to accommodate various on-board duplexers for multiple supported frequency bands without needing any external matching components. It achieves a gain higher than 21 dB, noise figure (NF) less than 2.65 dB and input referred third order intercept point (IIP3) value higher than 0 dBm in the frequency range of 0.7-2.7 GHz. This design is implemented in 65nm CMOS technology. It occupies a total area of 0.14 mm2 and consumes 4.5 mW from a 1.5 V supply
Life cycle assessment and optimization-based decision analysis of construction waste recycling for a LEED-certified university building
No full textThe current waste management literature lacks a comprehensive LCA of the recycling of construction materials that considers both process and supply chain-related impacts as a whole. Furthermore, an optimization-based decision support framework has not been also addressed in any work, which provides a quantifiable understanding about the potential savings and implications associated with recycling of construction materials from a life cycle perspective. The aim of this research is to present a multi-criteria optimization model, which is developed to propose economically-sound and environmentally-benign construction waste management strategies for a LEED-certified university building. First, an economic input-output-based hybrid life cycle assessment model is built to quantify the total environmental impacts of various waste management options: recycling, conventional landfilling and incineration. After quantifying the net environmental pressures associated with these waste treatment alternatives, a compromise programming model is utilized to determine the optimal recycling strategy considering environmental and economic impacts, simultaneously. The analysis results show that recycling of ferrous and non-ferrous metals significantly contributed to reductions in the total carbon footprint of waste management. On the other hand, recycling of asphalt and concrete increased the overall carbon footprint due to high fuel consumption and emissions during the crushing process. Based on the multi-criteria optimization results, 100% recycling of ferrous and non-ferrous metals, cardboard, plastic and glass is suggested to maximize the environmental and economic savings, simultaneously. We believe that the results of this research will facilitate better decision making in treating construction and debris waste for LEED-certified green buildings by combining the results of environmental LCA with multi-objective optimization modeling
Power-aware delay test quality optimization for multiple frequency domains
No full textAs the number of frequency domains aggressively grows in today's systems-on-chip (SoCs), the delivery of high-delay test quality across numerous frequency domains while meeting test budgets assumes crucial importance. This paper proposes a method to explore the delay test quality tradeoffs across these domains, determining an optimal distribution of the test time budget across all domains while minimizing the overall SoC delay defect escape level. Satisfaction of this goal necessitates not only consideration of fault coverage but also of t.e distinct characteristics of each domain, such as frequency, path length distribution, scan length, and shift speed as well as full utilization of concurrent test support while remaining within the constraints of power thresholds to provide a reliable test environment. An optimization formulation as well as efficient test time allocation methods based on convexity and fast concurrent test planning algorithms are provided
A hybrid HMM model for travel path inference with sparse GPS samples
Get PDFIn this study, we propose a novel method for a travel path inference problem from sparse GPS trajectory data. This problem involves localization of GPS samples on a road network and reconstruction of the path that a driver might have been following from a low rate of sampled GPS observations. Particularly, we model travel path inference as an optimization problem in both the spatial and temporal domains and propose a novel hybrid hidden Markov model (HMM) that uses a uniform cost search (UCS)-like novel combinational algorithm. We provide the following improvements over the previous studies that use HMM-based methods: (1) for travel path inference between matched GPS positions, the proposed hybrid HMM algorithm evaluates all candidate paths to find the most likely path for both the temporal and spatial domains. In contrast, previous studies either create interpolated trajectories or connect matched GPS positions using the shortest path assumption, which might not be true, especially in urban road networks (Goh et al. 2012; Lou et al. 2009). (2) The proposed algorithm uses legal speed limits for the evaluation of discrepancy in the temporal domain as in Goh et al. (2012), and Lou et al. (2009) only if there is not sufficient historical average speed data; otherwise, we use historical average speed computed from data. Our experiments with real datasets show that our algorithm performs better than the state of the art VTrack algorithm (Thiagarajan et al. 2009), especially for cases where GPS data is sampled infrequently