Yazar "Efeoğlu, Ebru" seçeneğine göre listele

Listeleniyor 1 - 6 / 6
  • Küçük Resim Yok
    Öğe
    Analysis of the Parameters that Affect the Measurements of Reflection Coefficients and Evaluation of the Effects of Parameters for K Nearest Neighbors-Based Liquid Classification
    (2021) Efeoğlu, Ebru; Tuna, Gürkan
    In this study, microwave spectroscopy method has been used in liquid measurements and K nearest neighbors algorithm has been used for classifying liquids. For this aim, firstly an experimental setup consisting of a vector network analyzer, a patch antenna and a bottle have been built to measure the reflection parameter of each liquid used in classification experiments. The aim of this study is to examine both the parameters that may affect the measurements taken with the proposed system and the algorithm parameters that may affect the performance in the classification of liquids and the effects of these parameters. Measurements have been taken by leaving different distances between the antenna and the liquid in order to examine whether the distance of the liquids to the antenna affects the measurement result, and if so, what effect. For examining the parameters of K nearest neighbors algorithm that may affect the classification, the scattering parameters of different liquids measured using the patch antenna have been used as microwave dataset. In addition, the effect of container type has been analyzed. Performance tests have been conducted by weighting and without weighting the algorithm, by measuring the accuracy rate when different numbers of nearest neighbors and different distance metrics have been used. The results reveal that the classification made by applying weighting is more successful than the classification made without weighting regardless of the number of nearest neighbors and used distance metrics.
  • Küçük Resim Yok
    Öğe
    CLASSIFICATION OF LIQUIDS USING A PATCH ANTENNA AND HIERARCHICAL CLUSTERING ALGORITHMS
    (2021) Efeoğlu, Ebru; Tuna, Gürkan
    Detection of hazardous liquids used in explosive production is important in terms of public safety and health. Because many threats can be prevented by detecting these liquids at security controls points. As existing methods have some disadvantages in terms of accuracy, practicality or reliability, there is a demand for new methods for hazardous liquid detection. In this paper, a circular patch antenna for hazardous liquid detection was designed and by connecting to a vector network analyzer a group of measurements was made. Then, this dataset was used by hierarchical clustering algorithms employed in this study to detect hazardous liquids. The results show that high classification accuracy can be achieved when Ward linkage method is preferred.
  • Küçük Resim
    Öğe
    Mikrodalga ölçüm verilerinden makine öğrenmesi teknikleri kullanılarak patlayıcı madde tespiti
    (Trakya Üniversitesi Fen Bilimleri Enstitüsü, 2021) Efeoğlu, Ebru; Tuna, Gürkan
    Tehlikeli sıvıların tespit edilmesi insan ve çevre sağlığının korunması açısından önemlidir. Çünkü bazı kimyasal sıvılar patlayıcı yapımında kullanılmakta ve bu patlayıcılarla teröristler saldırılar yapmaktadırlar. Terör saldırılarını önlemek için bu sıvıların diğer sıvılardan ayırt edilmesi gerekir. Bu nedenle araştırmacılar sıvı kontrolü yapan sistemler geliştirmek için çeşitli çalışmalar yürütmüşlerdir. Patlayıcı yapımında kullanılan sıvılar geliştirilen sıvı tarama sistemleri ile tespit edilmeye çalışılmıştır. Şuan kullanılmakta olan sıvı tarama sistemlerindeki yanlış alarm oranının yüksek olması ve bazı tehlikeli sıvıların tespit edilememesi araştırmacıları farklı yöntemler arayışına yöneltmiştir. Ayrıca operatöre bağlı sıvı tarama sistemlerinde operatörün dikkati kısa sürede dağılabilir, bu da sonuçların güvenilirliğini azaltabilir. Bu nedenle havalimanları, tren istasyonları, alışveriş merkezleri gibi insan nüfusunun kalabalık olduğu yerlerde bu tür saldırıları önlemek için sıvı kapağının açılması gerekmeyen ve yüksek doğrulukta sonuçlar veren sıvı tarama sistemleri kullanılmalıdır. Bu tezde mikrodalga ve makine öğrenmesi tabanlı bir sıvı tanıma sistemi önerilmektedir. Önerilen sistem sıvı ve akışkanların mikrodalga frekans bandındaki elektromanyetik tepki ölçümlerine dayanmaktadır. Bu sistem hızlı, tahribatsız, operatöre bağımsız ve yüksek doğrulukla sonuç vermektedir. Ayrıca, bu sistem %70 ve daha fazla tehlikeli sıvı konsantrasyonuna sahip sıvıları tespit edebilmektedir. Çünkü çözeltilerde %70 ve üzeri bir sıvı konsantrasyonu da tehlike yaratabilmektedir. Önerilen sistem hem tehlikeli sıvıların tehlikesiz sıvılar arasından tespiti hem de bu tehlikeli sıvıların sulu v çözeltilerindeki tehlikeli sıvı konsantrasyonunu bulmak için kullanılmıştır. Her iki uygulamada en başarılı algoritmayı bulmak için kullanılan algoritmaların performans analizi yapılmıştır. Analizde yedi farklı algoritma kullanılmış ve bu algoritmalar arasından en iyi performansı K-EYK algoritması göstermiştir.
  • Küçük Resim Yok
    Öğe
    A New Approach for Liquid Scanners to Determine Flammable Liquid Concentration in Solutions
    (2021) Efeoğlu, Ebru; Tuna, Gürkan
    Strong liquid explosives were obtained by mixing some chemical liquids and these explosives were used in many terrorist attacks in crowded places such as airports, railway stations and shopping malls. They were also used to cause sabotage to facilities that produce, store or use hazardous chemicals in their processes. For this reason, it is very important to take the necessary measures to prevent sabotage and terrorist attacks that may occur in such places in order to ensure public and environmental safety. In this study, a machine learning based liquid control system that can be used in airports, railway stations and shopping malls as well as in places with high fire probability is proposed. The difference of the proposed system from traditional liquid scanner systems is that it can detect the hazardous liquid concentration in the solutions as well as the detection of pure flammable liquids. Linear Discriminant Analysis and Quadratic Discriminant Analysis are used as classifiers and the performances of these techniques are compared. The results show that Quadratic Discriminant Analysis offers higher accuracy and lower error rates compared to Linear Discriminant Analysis.
  • Küçük Resim Yok
    Öğe
    Performance Evaluation of Sequential Minimal Optimization and K* Algorithms for Predicting Burst Header Packet Flooding Attacks on Optical Burst Switching Networks
    (2021) Efeoğlu, Ebru; Tuna, Gürkan
    Optical burst switching networks are vulnerable to various threats including Burst Header Packet Flooding attack, Circulating Burst Header attack, Address Spoofing, and Replay attack. Therefore, detecting such threats play a key role in taking appropriate security measures. One of the major challenges in identifying the risks of Burst Header Packet flooding attacks is the lack or insufficiency of reliable historical data. In this paper, firstly, Burst Header Packet flooding attacks are classified into four categories, Misbehaving-Block, Behaving-No Block, Misbehaving-No Block and Misbehaving-Wait, using Naive Bayes and K-Nearest Neighbor algorithms. Using performance metrics obtained both after testing on the same set and after applying 10-fold cross validation, the performance of Naive Bayes and K-Nearest Neighbor algorithms is compared based on commonly used performance metrics. As the results show, compared to Naive Bayes, K-Nearest Neighbor algorithm is more suitable for predicting Burst Header Packet Flooding attacks.
  • Küçük Resim Yok
    Öğe
    Use of Reflection Coefficients and Decision Tree Algorithm for Rapid Classification of Hazardous Chemical Liquids
    (2022) Efeoğlu, Ebru; Tuna, Gürkan
    The purpose of occupational health and safety studies is to protect employees from work accidents and occupational diseases and to ensure that they work in a healthy environment. Most of the work accidents happen as a result of wrong storage, transportation and use of chemicals. In order to protect employees from chemical hazards and eliminate their possible risks, a risk assessment should first be carried out. Based on the results of the risk assessment, if not done before, bottles that contain hazardous chemicals must be classified and labelled according to their risk levels. The labels of bottles that contain chemical liquids must be checked and if the labels are worn or unreadable, they must be renewed. After these have been done, hazardous chemical liquids must be classified, stored and transported according to these labels. In this study, a non-contact, liquid measurement system based on microwave data is proposed to detect hazardous liquids. In order to select the most suitable algorithm for use in this measurement system, 3 different classification algorithms have been used and the performance analysis of the algorithms has made. In the study, 3 different classification processes have been applied according to the chemical properties of the liquids. It has been observed that Random Tree algorithm has achieved the best performance while Rep Tree algorithm has done the worst performance. Using this system, hazardous chemical liquids can be detected without opening the cover of the bottles that contain the liquids. Therefore, it can be used to quickly label hazardous liquids for their safe storage and transportation.