Search

Nowadays it is important to store sensitive data and restrict its usage only to authorized people or institutions. In general, solutions for Data Leakage Prevention (DLP) ignores malicious attacks on documents and algorithms using fingerprinting and regular expressions are used. However, content-based attacks are successful evading those algorithms. In this paper an algorithm robust against malicious content-based attacks is proposed, which is independent of the attack executed. Transposition, sentence structure alteration, modification, obfuscation attacks are taken into consideration within the scope of paper. N-gram, charactergram, k-skip-n-gram and LSA methods are used in the feature extraction step, for having better classification results under attacks. The extracted features are passed to a Vote Classifier consisting of Support Vector Machine, Random Forest and Multi-Layer Perceptron classifiers. Additionally, the effects of instrumenting Spell-Correction in different steps of the algorithm is evaluated, which is effective against modification attacks. [ABSTRACT FROM AUTHOR]

Güç sistemleri, üretim aşamasından tüketim aşamasına kadar karmaşık bir yapıdadır. Elektrik enerjisinin tüketiciye sürekli, güvenilir, kaliteli ve ekonomik olarak ulaştırılması üretim/iletim/dağıtım aşamalarında kesintisiz ve optimize edilmiş enerji akışının sağlanmasına bağlıdır. Optimum Yük Akışı (OYA) problemi jeneratörler/bara gerilimleri, kademe değiştiricileri aktif/reaktif gücü güvenli sınırlarında tutup, yakıt maliyeti ve aktif/reaktif güç kayıpları gibi parametrelerini optimize etmeyi hedefler. Bu nedenle güç sistemlerinde OYA probleminin etkili bir yöntemle optimize edilmesi önemlidir. Bu yayın, OYA problemini çözmek için yeni bir sezgisel algoritma olan İyileştirilmiş Su Çevrim Algoritması (İSÇA)'nı kullanmaktadır. Önerilen yöntem farklı amaç fonksiyonlarına sahip IEEE 30 bara güç sistemi üzerinde uygulanmıştır. Önerilen yöntemin etkinliği ve esnekliği literatürde yakın zamanda yayınlanmış diğer sezgisel yöntemlerin sayısal sonuçlarıyla karşılaştırılarak tartışılmıştır. Bu çalışmada İSÇA yöntemiyle IEEE 30 bara güç sistemine ait önemli sonuçlarından biri olan yakıt maliyetinin sonucu 798.9021 $/s hesaplanmıştır. Bu veri, çalışmada bildirilen güncel yöntemlere ait yakıt maliyeti sonuçlarından daha iyidir. Ayrıca bu yayında sunulan yöntem gerçek bir güç sistemi olan 154 kV Güney Marmara iletim sisteminin bir kesitinin sanal modeline uygulanarak, gerçek zamanlı bir sistemde aktif güç kayıplarının azaltılmasının sağlanabileceği gösterilmiştir. Power systems are getting more complex structure from production to consumption stages. The continuous, reliable, quality and economical delivery of electrical energy to the consumer depends on ensuring uninterrupted and optimized energy flow during the production/transmission/distribution stages. Optimum Power Flow (OPF) problem goals to optimize the parameters such as fuel cost, active/reactive power losses, etc by keeping the generators/bus voltages, tap setting and active/reactive power within their secure limits. Therefore, it is important to optimize load flow in power systems with an effective method. This paper uses a recent heuristic algorithm, named as the Modified Water Cycle Algorithm (MWCA), for the OPF problem. For the purpose demonstrate the effectiveness of the recommended method, it has been analyzed on the IEEE 30 bus power system for varied object functions. The validity and flexibility of this algorithm are discussed by comparison with numerical conclusion of the many newly published heuristic methods in the literature. In this study, the result of the of fuel cost, which is important results of the IEEE 30 bus test system, is computed using MWCA as $798.9021/s. This value is best the fuel cost results of various methods reported in the study. In addition, this method, has applied to a part of the virtual model of 154 kV South Marmara transmission system to show that the active power losses of transmission system could be reduced in a realtime system. [ABSTRACT FROM AUTHOR]

With the use of computers in every field, the usage intensity of processors is increasing day by day. In order to cope with this intensity, processors need to be constantly updated and improved. Various scheduling algorithms are being developed every day in order to continuously improve and accelerate the processors and to cope with the processing intensity. The main purpose of the scheduling algorithms is to increase the efficiency of the processor. For this, the processor needs to make a planning between the pending processes. Planning can be done by giving different priorities or using different methods. In order to increase the efficiency of the processor, in this paper, existing algorithms were examined. Round Robin Algorithm was improved using Genetic Algorithm. It is aimed to find the optimum values by selecting the quantum value according to the pending processes and their parameters. When the developed new method compared with the existing standard algorithms, it is observed that the waiting time and the turnaround time were shorter than the other algorithms. [ABSTRACT FROM AUTHOR]

This paper presents a filled function method to solve economic load dispatch (ELD) of the thermal unit. The proposed method can take care of ELD considering nonlinearity such as valve point loading. The effectiveness of the proposed algorithm has been verified on three different test systems with equality and inequality constraint. Simulation results show that the filled function method is less sensitive to initial values of parameters and is more effective than other previously developed algorithms. In addition, the proposed algorithm is highly promising for the large-scale system. [ABSTRACT FROM AUTHOR]

Nowadays, a lot of data hiding algorithms for binary images have been proposed in the literature. The main purpose of these algorithms is to provide data hiding with high capacity and high visual quality. The pattern based data hiding algorithms are commonly used to achieve this purpose. In this paper, a new data hiding algorithm is suggested. This algorithm is based on minesweeper game for binary images. Neighborhood degree is calculated by selecting mine value as 0 or 1. The size of 3x3 non overlapping blocks have been used to determine neighborhood degree. Data hiding process is implemented by modifying the obtained neighborhood degree. Hamming distance is calculated to keep distortion in minimum level. In the selection of similar blocks with different neighborhood degrees, privacy of the algorithm is provided by using random number generators. The experimental results showed that the proposed data hiding algorithm resulted successfully. [ABSTRACT FROM AUTHOR]

Recent research in architecture have focused on to understand computational methods of formal exploration and expression. Computational mechanisms have been used to readdress formal issues using new techniques and methods. Computational tools that operate on algorithmic logic are central in these explorations. Despite the proliferation of computers in architecture today, use of algorithms in architectural design is limited. This paper describes "Designing the Design" studio work, developed within the scope of "The integration of Computational Design into Design and Architectural Education" research project. The aim of the studio is to explore the use of algorithms in architectural design process. The paper is organized in three parts, the first part elaborates formal language and algorithmic approach to design. The second describes the studio work in which algorithmic logic is integrated into design process and the third evaluates the studio work and discusses the changing paradigms in practice and education of design. [ABSTRACT FROM AUTHOR]

This paper presents a high-performance vector control strategy for Multilevel Inverter (MLI) fed induction motor drive system. In the proposed drive system, an adaptive and online trained sugeno type neuro-fuzzy speed controller is used to improve the speed tracking performance. The generation of Pulse Width Modulation (PWM) patterns for multilevel inverter is obtained using space vector PWM algorithm. The complete drive system is implemented in real time using DS1104 controller card. Using the experimental setup, the performance of the proposed drive system is tested under different speed commands, mechanical parameter variations and load disturbances. The experimental results show that control performance of the proposed drive system is satisfactory. [ABSTRACT FROM AUTHOR]

In this paper, the different working structure and unique design of the E-Core Transverse Flux Machine that is combined of transverse flux and reluctance principle is investigated for training bike application. Also, static magnetic characteristics of the machine are obtained with finite element analysis and experimental measurements. The modeling of the E-Core Transverse Flux Machine and its driver system are given. The whole system with all control algorithms for a training bike application is simulated and the performance curves are presented. Finally, the experimental tests done at the laboratory for the indoor training bike application are compared and verified with the results of the simulation. [ABSTRACT FROM AUTHOR]

In this paper a solution method is proposed for emphasizing the importance of costumer needs at the starting of Quality Function Deployment (QFD) application when there are multi costumer groups. Data Envelopment Analysis (DEA) steps are added to an algorithm in which an Analytic Network Process (ANP) approach is used in QFD. The need for adding DEA steps is determining the efficient customer requirements when there are multi customer groups. This solution approach is called DEANP which aggregates data DEA to ANP. The relative importance values of product technical requirements are calculated with this algorithm and application is performed at a manufacturer of white goods company in Turkey. The fact that using DEANP instead of ANP produced more powerful solutions is also confirmed by marketing specialists. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new hybrid meta-heuristic algorithm based on simulated annealing supplemented with tabu list for the Multi Objective Facility Layout Problem (MOFLP). It is a SA approach supplemented with a tabu list. The main advantage of this approach is that a short term memory provided by the tabu list can be used to avoid solution re-visits while preserving the stochastic nature of the SA method. The goal of the new hybrid heuristic algorithm is to find the solutions forming the set of efficient solutions (pareto optimal set) of the MOFLP's in a relatively short time. The hybrid meta-heuristic is illustrated with a practical example of plant with 8 departments and 4 objectives. An efficient set of layouts corresponding to pre-determined weights are obtained. And also the set of efficient pareto-optimal solutions of the problem are determined. [ABSTRACT FROM AUTHOR]

Wireless Sensor Networks (WSN) are very popular area of interest both in academic studies and industrial applications since they can work individually without additional maintenance and can be deployed in diverse applications. However, the limited source of the nodes cannot guarantee a sufficient level of security and also complicates the design and the implementation of the algorithms/protocols. DoS (Denial of Service) attacks that manipulate vulnerabilities of access protocol by disrupting network communication or arising exceptional cases and cutting network's lifetime are primary threats for wireless sensor networks. S-MAC (Sensor Medium Access Control) and its derivatives (T-MAC [1], D-MAC [2] etc.) which, are the most common medium access protocols designed for WSNs, are vulnerable against such attacks. In this paper, we have implemented the design of the AR-MAC (Attack Resistant MAC) protocol that detects the types of jammers and maintains the appropriate solution for each jammer type. By means of the new MAC protocol designed, the WSNs have been more robust to DoS attacks in MAC layer and the lifetime of the nodes have been extended under such attacks without additional hardware. [ABSTRACT FROM AUTHOR]