Yazar "Ünver, Halil Murat" seçeneğine göre listele
Listeleniyor 1 - 20 / 24
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A New Hybrid Approach Using GWO and MFO Algorithms to Detect Network Attack(Tech Science Press, 2023) Dalmaz, Hasan; Erdal, Erdal; Ünver, Halil MuratThis paper addresses the urgent need to detect network security attacks, which have increased significantly in recent years, with high accuracy and avoid the adverse effects of these attacks. The intrusion detection system should respond seamlessly to attack patterns and approaches. The use of metaheuristic algorithms in attack detection can produce near-optimal solutions with low computational costs. To achieve better performance of these algorithms and further improve the results, hybridization of algorithms can be used, which leads to more successful results. Nowadays, many studies are conducted on this topic. In this study, a new hybrid approach using Gray Wolf Optimizer (GWO) and Moth-Flame Optimization (MFO) algorithms was developed and applied to widely used data sets such as NSL-KDD, UNSW-NB15, and CIC IDS 2017, as well as various benchmark functions. The ease of hybridization of the GWO algorithm, its simplicity, its ability to perform global optimal search, and the success of the MFO algorithm in obtaining the best solution suggested that an effective solution would be obtained by combining these two algorithms. For these reasons, the developed hybrid algorithm aims to achieve better results by using the good aspects of both the GWO algorithm and the MFO algorithm. In reviewing the results, it was found that a high level of success was achieved in the benchmark functions. It achieved better results in 12 of the 13 benchmark functions compared. In addition, the success rates obtained according to the evaluation criteria in the different data sets are also remarkable. Comparing the 97.4%, 98.3%, and 99.2% classification accuracy results obtained in the NSL-KDD, UNSW-NB15, and CIC IDS 2017 data sets with the studies in the literature, they seem to be quite successful.Öğe A review on the evolution of induction fluid heaters(Taylor & Francis Inc, 2022) Keleşoğlu, Alper; Kanmaz, Nergiz; Ünver, Halil Murat; Ünver, ÜmitFossil fuel firing heating systems will be substantially abandoned in the near future due to the zero-carbon goals. They will be replaced with heating systems that use renewable energy sources. Since the investment costs and area requirements of the renewable energy systems are high, the developments of highly efficient electric fluid heating systems have critical importance for a sustainable, environmentally friendly and clean heating. Among the electrical heating systems, induction heaters offer quite easy to use and better adaptability to most heating systems by allowing fluid heating without the necessity for contact. In this paper, a comprehensive review of the evolution of the induction fluid heaters is given by presenting the historical background. The most important milestones of induction fluid heating systems are presented. The development strategies suggested by the different papers including patents are investigated. The findings revealed that induction fluid heating systems, which reach approximately 100% thermal efficiency. These systems are introduced as follows: clean, non-emitting fluid heating system. Because of this, the study concludes that induction fluid heaters are one of the most promising alternatives to other electric heating systems in the long run.Öğe An Attack Detection Framework Based on BERT and Deep Learning(IEEE-Inst Electrical Electronics Engineers Inc, 2022) Seyyar, Yunus Emre; Yavuz, Ali Gökhan; Ünver, Halil MuratDeep Learning (DL) and Natural Language Processing (NLP) techniques are improving and enriching with a rapid pace. Furthermore, we witness that the use of web applications is increasing in almost every direction in parallel with the related technologies. Web applications encompass a wide array of use cases utilizing personal, financial, defense, and political information (e.g., wikileaks incident). Indeed, to access and to manipulate such information are among the primary goals of attackers. Thus, vulnerability of the information targeted by adversaries is a vital problem and if such information is captured then the consequences can be devastating, which can, potentially, become national security risks in the extreme cases. In this study, as a remedy to this problem, we propose a novel model that is capable of distinguishing normal HTTP requests and anomalous HTTP requests. Our model employs NLP techniques, Bidirectional Encoder Representations from Transformers (BERT) model, and DL techniques. Our experimental results reveal that the proposed approach achieves a success rate over 99.98% and an F1 score over 98.70% in the classification of anomalous and normal requests. Furthermore, web attack detection time of our model is significantly lower (i.e., 0.4 ms) than the other approaches presented in the literature.Öğe Öğe Control of A Car, Via Internet(Kırıkkale Üniversitesi, 2018) Erbir, Muhammed Ali; Ünver, Halil MuratInternet and mobile devices such as phones, tablets, and even wearable goods are getting more popular day by day. Embedded system technology offers fast, cheap, and optimized solutions for designers with its wide portfolio. This study demonstrates a remote-control application of a car via internet platform. An embedded system was designed with an Arduino card to interact with the car hardware and an android application was developed to control the car. The car was controlled by the android application via an internet connected embedded system successfully.Öğe Design of a Fuzzy Logic Based Custom Exam Production System for High Performance(Kırıkkale Üniversitesi, 2020) Taşkırdı, Özkan; Ünver, Halil MuratIn this study, a software which aims to keep the performance and motivation at the highest level by reducing the psychological effects of positive / negative differences in students' performance has been developed by using Fuzzy Logic infrastructure in C # language. In the study based on the mathematics course, the number of correct answers given to the unit questions of the students preparing for the university entrance exam, the number of correct answers of the other students taking the examination and the passing grades of the courses taken in the previous years were evaluated with fuzzy logic and the learning rates of the students were determined for each unit. According to the determined learning rates, exams with question combinations suitable for the level of the students were derived. With the developed software, it is aimed to achieve a continuous development line based on keeping the students' morale-motivation level high. It was observed that work motivation and student achievement were significantly increased with test studies conducted with specially derived questions.Öğe Detection of tibial fractures in cats and dogs with deep learning(Ankara Univ Press, 2021) Baydan, Berker; Ünver, Halil MuratThe aim of this study is to classify tibia (fracture/no fracture) on whole/partial body digital images of cats and dogs, and to localize the fracture on fracture tibia by using deep learning methods. This study provides to diagnose fracture on tibia more accurately, quickly and safe for clinicians. In this study, a total of 1488 dog and cat images that were obtained from universities and institutions were used. Three different studies were implemented to detect fracture tibia. In the first phase of the first study, tibia was classified automatically as fracture or no fracture with Mask R-CNN. In the second phase, the fracture location in the fracture tibia image that obtained from the first phase was localized with Mask R-CNN. In the second study, the fracture location was directly localized with Mask R-CNN. In the third study, fracture location in the fracture tibia that obtained from the first phase of first study was localized with SSD. The accuracy and F1 score values in first phase of first study were 74% and 85%, respectively and F1 score value in second phase of first study was 84.5%. The accuracy and F1 score of second study were 52.1% and 68.5%, respectively. The F1 score of third study was 46.2%. The results of the research showed that the first study was promising for detection of fractures in the tibia and the dissemination of the fracture diagnosis with the help of such smart systems would also be beneficial for animal welfare.Öğe Detection of Web Attacks Using the BERT Model(IEEE, 2022) Seyyar, Yunus Emre; Yavuz, Ali Gökhan; Ünver, Halil MuratThis paper presents a web intrusion detection system that addresses security threats with the increasing use of web applications in almost all domains, as well as the increase in attacks against web applications. Our web intrusion detection system consists of a model that can distinguish between normal and abnormal URLs. In the URL analysis phase, our model uses the BERT model of Transformers, a prominent natural language processing technique. In the classification phase, we use a CNN model, which is a popular deep learning technique. We utilize the CSIC 2010, FWAF, and HttpParams datasets for training and testing. The experimental results show that our model performs the classification of normal and abnormal requests in 0.4 ms, which is an extremely fast detection time when compared to the reported results in the literature and an accuracy of over 96%.Öğe Determining the Location of Tibial Fracture of Dog and Cat Using Hybridized Mask R-CNN Architecture(Kafkas Univ, Veteriner Fakultesi Dergisi, 2021) Baydan, Berker; Barışçı, Necaattin; Ünver, Halil MuratThe aim of this study is to hybridize the original backbone structure used in the Mask R-CNN framework, and to detect fracture location in dog and cat tibia fractures faster and with higher performance. With the hybrid study, it will be ensured that veterinarians help diagnose fractures on the tibia with higher accuracy by using a computerized system. In this study, a total of 518 dog and cat fracture tibia images that obtained from universities and institutions were used. F1 score value of this study on total dataset was found to be 85.8%. F1 score value of this study on dog dataset was found to be 87.8%. F1 score value of this study on cat dataset was found to be 77.7%. With the developed hybrid system, it was determined that the localization of the fracture in an average tibia image took 2.88 seconds. The results of the study showed that the hybrid system developed would be beneficial in terms of protecting animal health by making more successful and faster detections than the original Mask R-CNN architecture.Öğe F KLAVYE İÇİN METİN ANALİZİ TABANLI KELİME TAMAMLAMA SİSTEMİ(2020) Karabulut, Bergen; Ünver, Halil MuratKelime tamamlama, kullanıcının basması gereken tuş sayısını azaltmayı hedefleyenbir yardımcı teknoloji aracıdır. İlk çalışmalarda, genellikle bir alternatif vedestekleyici iletişim aracı olarak ele alınmıştır. Fakat son dönemlerde kelimetamamlama sistemleri; mobil cihazlar, makine çevirisi ve arama motorları gibi farklıalanlarda da önem kazanmıştır. Bu çalışmada, Standart Türkçe Klavye olan F klavyeiçin metin analizi tabanlı bir kelime tamamlama sistemi geliştirilmiştir. Geliştirilensistemi test etmek için zabıt kâtiplerinin yoğun klavye kullanımından dolayı AdaletBakanlığı tarafından yapılan zabıt kâtipliği sınavlarına ait 160 metin kullanılmıştır.Bu metinlerdeki kelimelerin unigram ve bigram frekansları çıkarılmış ve birveritabanında tutulmuştur. Sistem, bu veritabanını kullanarak yazma işlemiesnasında kullanıcıya 8 adet alternatif kelimeden oluşan bir tahmin listesisunmaktadır. Tanımlanan tuş kombinasyonları yardımıyla kullanıcı tahminlistesinden bir kelime seçebilmekte ve seçilen kelime sistem tarafından otomatikolarak tamamlanmaktadır. Geliştirilen sistemin performansı tuşlama tasarrufuaçısından değerlendirilmiştir. Tüm metinler arasından rastgele seçilen 15 metin ileyapılan test işleminde ortalama %50,98 tuşlama tasarrufu sağlanmıştır.Öğe Hipertansiyon Tahmini İçin Temel Bileşen Analizinin Kullanımı(2020) Ünver, Halil Murat; Kökver, Yunus; Çiftçi, AydınAmaç: Otuz yaş ve üzerindeki 150 hastadan, hipertansiyona etki etmesi muhtemel bilgilerden; cinsiyet, yaş, lipid profili, trigliserid, vücut kütle indeksi, ürik asit ve sigara kullanımı verileri toplanmış ve bir hipertansiyon veritabanı oluşturulmuştur. Bu kişilerden 65’i sağlıklı, geriye kalan 85 kişi ise hipertansiyon hastasıdır. Bu veritabanından hipertansiyon hastalığının Temel Bileşen Analizi kullanılarak tahmin edilmesi amaçlanmıştır.Gereç ve Yöntem: Naive Bayes, Çok Katmanlı Algılayıcı Ağ (ÇKA), Karar Tablosu ve C4.5 sınıflandırma algoritmaları uygulanmış, ardından Temel Bileşenler Analizi uygulanarak hipertansiyon veritabanının boyutu indirgenmiş ve aynı sınıflandırma algoritmaları tekrar uygulanmış ve sonuçlar karşılaştırılmıştır.Bulgular: Aynı şartlarda işleme sokulan algoritmalardan en başarılı sonucu %88 doğruluk oranıyla Naive Bayes sınıflandırıcısı vermiştir. Naive Bayes sınıflandırıcısını sırasıyla %85,33 başarı oranıyla Karar Tablosu algoritması, %82,67 başarı oranıyla ÇKA algoritmaları takip etmiştir. Hipertansiyon veritabanına TBA analizi uygulanıp, aynı şartlarda aynı algoritmalar tekrar işleme sokulup, TBA uygulanmayan sonuçlarla kıyaslandığında ise, C4.5 algoritması normalden %4 daha başarılı sonuç vererek en başarılı algoritma olmuştur. C4.5 algoritmasını sırasıyla %2,67 daha başarılı sonuç veren Karar Tablosu algoritması ve %1,33 daha başarılı sonuç veren ÇKA izlemiştir.Sonuç: Naive Bayes sınıflandırıcı haricindeki tüm algoritmalarda Temel Bileşenler Analizi’nin sınıflandırma başarısını artırdığı görülmüştür.Öğe İndüksiyonlu çelik tav fırınlarında güç ünitelerinin PLC ile denetimi(Kırıkkale Üniversitesi, 2004) Ünver, Halil Murat; Çelik, Veli; Aydemir, Y.Mehmet TimurÖZET İNDÜKSİYONLU ÇELİK TAV FIRINLARINDA GÜÇ ÜNİTELERİNİN PLC İLE DENETİMİ ÜNVER, Halil Murat Kırıkkale Üniversitesi Fen Bilimleri Enstitüsü Makine Anabilim Dalı, Doktora Tezi Danışman : Prof. Dr. Veli Çelik Ortak Danışman : Yrd. Doç. Dr. Mehmet Timur Aydemir Ocak 2004, 122 Sayfa İndüksiyonlu çelik tav fırınlarında malzeme üzerinde oluşan ısı, manyetik geçirgenlik katsayısının düşmesine, elektriksel direncin ise artmasına neden olmaktadır. Parametrelerdeki bu değişim, elektrik şebekesinden çekilen aktif gücün azalması ve istenmeyen reaktif gücün artması anlamına gelmektedir, ve dolayısıyla malzeme üzerindeki sıcaklık arttıkça enerji transfer oranı düşmektedir. Ayrıca, güç ve frekans ayarlı indüksiyon fırınlarında oluşturulan yapılar, uygun bir veri iletişim alt yapısı sağlamadığından modern kontrol tekniklerinin uygulanması da mümkün olmamaktadır. Yeni kuşak PLC'Ier, indüksiyon fırınlarının mal yükleme, ısı, basınç, akım, gerilim kontrol-kumandasını yapabilecek yeterliliktedirler. Ayrıca, buPLC'Ier indüksiyon fırınlarının güç ünitesi kısmını oluşturan eviricilerin frekans ve güç kontrolünü yapabilecek kapasiteye de sahiptir. Bu tez çalışmasında, yeni kuşak PLC'Ierin bu özellikleri göz önüne alınarak, indüksiyonlu çelik tav fırınlarında kullanımı öngörülmüştür. Böylelikle, indüksiyonlu çelik tav fırınlarında modern kontrol tekniklerinin uygulanması için bir alt yapı oluşturulması amaçlanmıştır. Tasarlanan deney düzeneği ile yapılan deneysel çalışmada, sistem, ısıtılan malzeme üzerinde meydana gelen manyetik iletkenlik ve özdirenç değişimlerini istenilen hassasiyette izlemiş, bu sayede, şebekeden sürekli aktif ve sabit seviyeli enerji çekilmesi gerçekleştirilmiştir. Ayrıca, PLC'nin yeterli düzeyde performans göstermesi sayesinde indüksiyonlu çelik tav fırını üzerinde modem kontrol tekniklerinin uygulanabilmesi için gerekli alt yapı oluşturulmuştur. Sistemin simülasyonu için yapılan çalışmalarda ise, MATLAB® Simulink programı üzerinde, deneysel düzeneğe ilişkin model oluşturulmuş ve simülasyon gerçekleştirilmiştir. Geliştirilen rezonans frekansı belirleme mantığının, simülasyon ortamında da deneysel çalışmalarda olduğu gibi güvenilir olduğu gözlenmiştir. ORCAD® Pspice programı ile yapılan simülasyonda elde edilen eğrilerle, MATLAB® Simulink'de kurulan modelde elde edilen eğrilerin bir karşılaştırması yapılmış, Simulink'de oluşturulan modelin tamamıyla güvenilir olduğu ortaya konmuştur. Anahtar Kelimeler: PLC, indüksiyonlu ısıtma fırınları, frekans ve güç kontrolü, güç ünitesi, eviriciÖğe KUZEM LMS: A new learning management system for online education(Sila Science, 2012) Ergüzen, Atilla; Erel, Şerafettin; Uzun, İbrahim; Bilge, Hasan Şakir; Ünver, Halil MuratE-learning has made rapid strides in recent years thanks to the advances in educational technology. Many educational institutions have started e-learning. In e-learning, there is a need for administration, documentation, delivery of the course content and tracking students' performances all of which are done by the help of a software called Learning Management System (LMS). A robust LMS is a requirement for achieving the goals of the institution providing e-learning. There are basically three kinds of EMS: open source, commercial and institution-based (self-developed). All of them have advantages and disadvantages. Choosing the type of LMS to be used by any institution depends on the decision makers of that institution. Decision makers in Kirikkale University chose to develop its own LMS depending on the individual needs of its distance education center. In this study, a new learning management system called KUZEM LMS, created by the software development team of Kirikkale University and used in the Distance Education Center of the same university is aimed to be introduced. The features of the KUZEM EMS are given in detail and a comparison of it is made with the other LMSs.Öğe Machine Learning Approaches in Detecting Network Attacks(Institute of Electrical and Electronics Engineers Inc., 2021) Dalmaz, Hasan; Erdal, Erdal; Ünver, Halil MuratDeveloping technology brings many risk in terms of data security. In this regard, it is an important issue to detect attacks for network security. Intrusion detection systems developed due to technological developlments and increasing attack diversity have revealed the necessity of being more succesful in detecting attacks. Today, many studies are carried out on this subject. When the literature is examined, there are various studies with varying success rates in detecting network attacks using machine learning approaches. In this study, the NSL-KDD dataset was explained in detail, the positive aspects of the KDD Cup 99 dataset were specified, the classifier used, performance criteria and the success results obtained were evaluated. In addition, the developed GWO-MFO hybrid algorithm is mentioned and the result is shared. © 2021 IEEEÖğe A new induction water heating system design for domestic heating(Sila Science, 2012) Ünver, Halil MuratIn the Middle Eastern and North African countries (between N36-N26 latitudes) where temperatures do not drop heavily in winter times, the resistance-boilers connected to the solar panels in serial offer a practical and economical solution for domestic heating. In this way, while solar power is benefited at the highest level, the desired room temperature can be achieved easily through the boiler having low resistance power. However, due to failure of resistances at short intervals and the need of laborious effort during the changing process, reduce the appeal of resistance-boilers. At this point, the boiler having a low breakdown probability and heating the water with induction heating principle offer a quite attractive solution. The aim of this study is to research induction water heating technique for electrical boiler applications. To this end, a special induction water-heating system was designed and produced. The designed system was run by single and two-phase electrical connection and satisfactory results were obtained. It is expected that an efficient heating system having low-cost operation and maintenance can be developed by improving this technique applicable for market.Öğe Paralel Rezonans Devrelerinde Basit ve Güvenilir Rezonans Frekansı Belirlemede Yeni Bir Yaklaşım(Kırıkkale Üniversitesi, 2010) Ünver, Halil MuratToday, parallel resonance circuits have been used in many fields. In the circuit structures where the induction values change, to determine whether the worked frequency is below or above the resonance frequency requires to perform complex circuit designs. The problems, faced during the improvement of determination and response of these complex circuits, decrease the reliability of such approaches. Since the inductance follows non-linear lines, as it is seen especially in steel heating and melting, the high noise effects originated from the switching bring out several usage problems in steel heating and melting. In this study, what we try to do is to obtain a simple and reliable resonance frequency determining techniques for the 60 kW induction steel heating furnace designed beforehand. The aimed technique has been tested on the MATLAB model system and has been used on the experimental set up in the light of obtained results.Öğe Principal Component Analysis Using For Estimating Hypertension(Kırıkkale Üniversitesi, 2020) Ünver, Halil Murat; Kökver, Yunus; Çifci, AydınAim: 150 patients which aged 30 years and over were exposed to possible hypertension; age, gender, lipid profile, body massindex, triglyceride , cigarette use and uric acid data are collected and hypertension database are created. 65 people is healthy, andthe remaining one is suffering from hypertension. It is aimed to estimate the hypertension disease from this database using thePrincipal Component Analysis.Material and Method: Decision Table, Naive Bayes, C4.5 and Multilayer Perceptron Network(MLP) classification algorithmsare applied to this database, then the size of the hypertension database is reduced by applying Principal Component Analysis andthe same methods are applied again and the results are compared.Results: The most successful result of the algorithms that were processed under the same conditions gave Naive Bayes classifierwith 88% accuracy. Naive Bayes classifier was followed by the Decision Table algorithm with success rate of 85.33%, and ÇKAalgorithms with success rate of 82.67%. If the TBA analysis is applied to the hypertension database and the same algorithms arere-processed under the same conditions and the TBA is compared to the untreated results, the C4.5 algorithm is normally themost successful algorithm with 4% more successful results. The Decision Table algorithm, which yielded C4.5 algorithm with2.67% more success rate respectively, and ÇKA which has a more successful result than 1.33%.Conclusion: Algorithms except the Naive Bayes algorithm, improved their classification accuracy rateÖğe Segmentation of blood vessels from retinal images(2017) Kökver, Yunus; Ünver, Halil Murat; Duman, Ebru AydoğanRetina görüntülerinden hastalık teşhisinin yapılabilmesinin ilk adımı kan damarlarının segmente edilmesidir. Bu çalışmada retina görüntüleri üzerinden kan damarlarının çıkartılması üzerine yapılan çalışmaları incelemeyi amaçlamaktadır. Bu nedenle literatürdeki mevcut makaleler kullanılan yöntemleri belirlemeye odaklanarak sistematik olarak derlenmiştir. Damar segmentasyonu problemine çözüm getiren ve literatürde bu alandaki ilk çalışmadan başlayarak son zamanlara kadar yapılan çalışmalardaki çözümler bazı kriterler dahilinde değerlendirilmiştir. Bu derleme çalışmasından anlaşılıyor ki, yıllar içerisinde segmentasyon için kullanılan yöntemlerde ciddi bir ilerleme kaydedilmiş ve retina görüntülerinden tüm damarların segmentasyonu kolaylıkla yapılabilir düzeye gelmiştir.Öğe The Do’s and Don’ts for Increasing the Accuracy of Face Recognition on VGGFace2 Dataset(Springer Science and Business Media Deutschland GmbH, 2021) Erbir, Muhammed Ali; Ünver, Halil MuratIn this study, developments in face recognition are examined. Some methods are presented to increase the accuracy rate in face recognition by using transfer learning with VGGFace2 dataset and 4 different CNN models. While some of these tested offers decreased the accuracy rate, some of them increased. Effects of histogram balancing, expanding the training data, extracting the effect of non-facial portions of images and vertically aligning images on the accuracy rate were determined and compared to the accuracy rates of original images. As the optimal solution, transfer learning from the InceptionV3 model was preferred, vertical positioning was made, and an accuracy rate of 95.47% was achieved when 10% of the images were used for testing and 90% for training in a 100 people subset of VGGFace2 dataset. In LFW, one of the widely used datasets in the literature, an accuracy rate of 100% has been achieved by exceeding the highest accuracy achieved so far and all images in the LFW database have been recognized without any problems. © 2021, King Fahd University of Petroleum & Minerals.Öğe Üniversite Öğrencilerinin Beslenme Alışkanlıklarının Bulanık Mantık Sistemi İle Değerlendirilmesi(Kırıkkale Üniversitesi, 2020) Taşkırdı, Özkan; Ünver, Halil MuratBu çalışma, rutin bir beslenme çevrimi oluşturamayan üniversite öğrencilerinde beslenmeye ilişkin alışkanlıkların ve bu alışkanlıkların oluşmasına neden olan etkenlerin ortaya çıkarılarak, dengeli beslenme alışkanlıklarının olup olmadığını belirlemek amacıyla yapılmıştır. Bu amaç doğrultusunda Ahi Evran Üniversitesi Kaman Meslek Yüksekokulu Gıda İşleme Bölümü öğrencilerine anket uygulanmış ve anket sonuçları kullanılarak bulanık mantık sistemi ile öğrencilerin beslenme şekilleri ortaya konulmuştur. Bu çalışmada Matlab programının Fuzzy toolbox’u kullanılarak Bulanık Mantık Sistemi oluşturulmuş ve sonucunda bulanık mantık sistemlerinin beslenme şeklinin değerlendirilmesin de tutarlı sonuçlar verdiği gözlemlenmiştir. İstatistiki olarak hesaplanan yöntemlere alternatif olarak beslenme şeklinin değerlendirmesi işlemlerinde ve Gıda Mühendisliği alanında bulanık mantığın kullanılması ile yeni yaklaşımlar geliştirilebilir.
