Person: ÜLKÜ, EYÜP EMRE
Loading...
Email Address
Birth Date
Research Projects
Organizational Units
Job Title
Last Name
ÜLKÜ
First Name
EYÜP EMRE
Name
15 results
Search Results
Now showing 1 - 10 of 15
Publication Open Access Classification of hazelnuts with CNN based deep learning system(2022-12-01) ÜLKÜ, EYÜP EMRE; YILDIZ, KAZIM; GÜNEŞ E., ÜLKÜ E. E., YILDIZ K.The rapid development of technology leads to the emergence of technology-based systems in many different areas. In recent years, agriculture has been one of these areas. We come across technological systems in agricultural applications for many different purposes such as growing healthier products, increasing the yield of products, and predicting product productivity. Today, technology-based systems are used more and more widely in agricultural applications. Classification of products quickly and with high accuracy is a very important process in predicting product yield. In this study, it is suggested to use the CNN-based deep learning model VGG16 in order to classify the hazelnut fruit, which is an important agricultural product. The main purpose is to classify hazelnuts according to their quality with a deep learning approach. For that, a new data set was created. There are 15770 images in the created data set. In the study, the data set was used by dividing it into different parts. The classification of hazelnut images was carried out using the VGG16 deep learning model, which is a powerful model for classifying images. As a result of the experiments on the data set created, the classification process of hazelnuts was realized with 0,9873 F1 score. The detection rate of quality hazelnut is 0.9848, the rate of detection of kernel hazelnut is 0.9891 and the rate of detection of damaged hazelnut is 0.9882. In addition, the classification process was carried out with deep learning using 50%, 25% and 10% of the data set in the study. It was observed that the 98.73 %, 95.46 %, 92.62 %, and 88.42 % accuracy rates were achieved when the whole, 50 %, 25 %, and 10 % data sets were used, respectivelyPublication Metadata only Forecasting greenhouse gas emissions based on different machine learning algorithms(2022-01-01) ÜLKÜ, EYÜP EMRE; Ulku I., ÜLKÜ E. E.© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.With the increase in greenhouse gas emissions, climate change is occurring in the atmosphere. Although the energy production for Turkey is increased at a high rate, the greenhouse gas emissions are still high currently. Problems that seem to be very complex can be predicted with different algorithms without difficulty. Due to fact that artificial intelligence is often included in the studies to evaluate the solution performance and make comparisons with the obtained solutions. In this study, machine learning algorithms are used to compare and predict greenhouse gas emissions. Carbon dioxide (CO2), nitrous oxide (N2O), methane (CH4), and fluorinated gases (F-gases) are considered direct greenhouse gases originating from the agriculture and waste sectors, energy, industrial processes, and product use, within the scope of greenhouse gas emission statistics. Compared to different machine learning methods, support vector machines can be considered an advantageous estimation method since they can generalize more details. On the other hand, the artificial neural network algorithm is one of the most commonly used machine learning algorithms in terms of classification, optimization, estimation, regression, and pattern tracking. From this point of view, this study aims to predict greenhouse gas emissions using artificial neural network algorithms and support vector machines by estimating CO2, CH4, N2O, and F-gases from greenhouse gases. The data set was obtained from the Turkish Statistical Institute and the years are included between 1990 and 2019. All analyzes were performed using MATLAB version 2019b software.Publication Metadata only A Performance Way Comparison of Docker Swarm and Kubernetes(2022-12-21) ÜLKÜ, EYÜP EMRE; Yurtsever O., ÜLKÜ E. E.People always try to find the best way to deploy their applications. Thus, the first solution was bare metal servers. In the bare metal server solution, each server was responsible for one application but that required as many servers as the number of applications, and in addition that this solution also required so much space in the server room. Subsequently, virtualization technology emerges. This technology is based on the abstraction of computer hardware. Virtualization technology enables us to host multiple operating systems in one host. Virtualization brings along with many advantages. One of the significant benefits is to reduce the quantity of physical equipment needed in the data center and helps to scale our applications. One of the remarkable developments in information technologies is container technology which emerged in the middle of 2010. Containers allow to package an application with all the parts it needs, such as libraries and other dependencies, and deliver it as a single package. Docker, developed by Google, is the essential tool to use in container technologies. Docker resembles a virtual machine, but as opposed to a virtual machine, instead of creating a whole virtual operating system, Docker allows applications to use the same Linux kernel as the system they use. This provides a performance boost and reduces the size of the application. Also, the development of microservice-based applications in recent years has made container technologies widely used. Now, we can run each of our applications, perhaps thousands of containers. However, this solution brings with it another problem, how do we manage these containers? In recent years, there have been a few breaks in the server side, which underlies the rapidly developing information technologies. The last of these is container technology. Before virtualization, companies were running all their applications on physical servers, and these systems were getting complex over time, making even simple problems inextricable. Accordingly, in this study we compare Docker Swarm and Kubernetes in terms of their performances under heavy load, two of the most used tools for container management. Thus, we aim to inform readers about container management.Publication Open Access du-CBA: veriden habersiz ve artırımlı sınıflandırmaya dayalı birliktelik kuralları çıkarma mimarisi(2023-01-01) BÜYÜKTANIR, BÜŞRA; YILDIZ, KAZIM; ÜLKÜ, EYÜP EMRE; Büyüktanır B., Yıldız K., Ülkü E. E., Büyüktanır T.İstemci sunucu sistemlerinde makine öğrenmesi modeli kullanılması bir ihtiyaçtır. Ancak istemcilerden verilerin toplanması, sunucuya aktarılması, makine öğrenmesi modeli eğitilmesi ve bu modelin istemcilerde çalışan cihazlara entegre edilmesi bir çok problemi beraberinde getirmektedir. Verilerin istemcilerden sunucuya transferi ağ trafiğine sebep olmakta, fazla enerji gerektirmekte ve veri mahremiyetini istismar edilebilmektedir. Çalışma kapsamında, bahsedilen problemlere çözüm için federe öğrenme mimarisi kullanılmaktadır. Mimariye göre, her bir istemcide istemcinin kendi verilerinden makine öğrenmesi modeli eğitilmektedir. Her bir istemcide eğitilen modeller sunucuya gönderilmekte ve sunucuda bu modeller birleştirilerek yeni bir model oluşturulmaktadır. Oluşturulan nihai model tekrar istemcilere dağıtılmaktadır. Bu çalışmada Veriden Habersiz İlişkili Kurallara Dayalı Sınıflandırma (Data Unaware Classification Based on Association, du-CBA) olarak adlandırılan ilişkisel sınıflandırma algoritması geliştirilmiştir. Federe öğrenme ile klasik öğrenme mimarilerini karşılaştırıp başarılarını ölçmek için çalışma kapsamında benzetim ortamı oluşturulmuştur. Benzetim ortamında du-CBA ve CBA algoritmaları kullanılarak modeller eğitilmiş ve sonuçlar kıyaslanmıştır. Modellerin eğitiminde University of California Irvine (UCI) veri havuzundan alınan beş veri seti kullanılmıştır. Deneysel sonuçlar, her bir veri seti için federe öğrenme ile eğitilen modellerin, klasik öğrenme ile eğitilen modellerle neredeyse aynı doğruluğu elde ettiğini ama eğitim sürelerinin yaklaşık %70 oranında azaldığını göstermiştir. Sonuçlar geliştirilen algoritmanın başarıya ulaştığını ortaya koymaktadır.Publication Open Access Sharing Location Information in Multi-UAV Systems by Common Channel Multi-Token Circulation Method in FANETs(KAUNAS UNIV TECHNOLOGY, 2019-02-12) DOĞAN, BUKET; Ulku, Eyup Emre; Dogan, Buket; Demir, Onder; Bekmezci, IlkerUnmanned Aerial Vehicle (UAV) technology is being used increasingly for military and civilian purposes. The primary reason for this increase is that UAVs eliminate the risk to human life in difficult and dangerous missions, are cost effective, and easily are deployed. Developments in UAV technology and decreasing costs have increased UAV usage. However, when multiple UAVs are deployed, inter UAV communication becomes complicated. For this reason, communication in multi-UAV systems is the most important problem that needs to be solved. To enable communication among UAVs without infrastructure support, a Flying Ad Hoc Network (FANET) is used. A FANET provides UAVs to fly in tandem without colliding. To ensure coordinated flight, UAVs require the location information of other UAVs. In this study, we developed a common channel multi-token circulation protocol to share location information in multi-UAV systems that communicate using a FANET. The proposed method ensures that UAVs in multi-UAV systems know each other's coordinate information with minimum error.Publication Open Access Fault detection of fabrics using image processing methods(PAMUKKALE UNIV, 2017) ÜLKÜ, EYÜP EMRE; Yildiz, Kazim; Demir, Onder; Ulku, Eyup EmreThis paper presents a computer aided detection (CAD) system which uses wiener filter based approach for detection of defects in poplin fabric. The defective fabric images are taken with the help of the digital camera. The developed system consists of three phases, including preprocessing, segmentation and detection of fabric defect In preprocessing phase, a RGB to gray level conversion and image enhancement operations were applied to digital camera images. In segmentation phase, background of the gray level image segmented using morphologic operations. Then, segmented image was converted to binary image to facilitate fabric defect detection process. Fabric defect detection was performed using wiener filter in the detection phase of the system. Wiener filter is applied to binary level image to eliminate structures which are not defect The developed detection system applied on defective poplin images for detection. The obtained results on different kinds of fabric defects show that the proposed algorithm gives promising results.Publication Metadata only Forecasting greenhouse gas emissions based on different machine learning algorithms(Springer, Cham, 2022-01-01) ÜLKÜ, EYÜP EMRE; ÜLKÜ İ., ÜLKÜ E. E.With the increase in greenhouse gas emissions, climate change is occurring in the atmosphere. Although the energy production for Turkey is increased at a high rate, the greenhouse gas emissions are still high currently. Problems that seem to be very complex can be predicted with different algorithms without difficulty. Due to fact that artificial intelligence is often included in the studies to evaluate the solution performance and make comparisons with the obtained solutions. In this study, machine learning algorithms are used to compare and predict greenhouse gas emissions. Carbon dioxide (CO2), nitrous oxide (N2O), methane (CH4), and fluorinated gases (F-gases) are considered direct greenhouse gases originating from the agriculture and waste sectors, energy, industrial processes, and product use, within the scope of greenhouse gas emission statistics. Compared to different machine learning methods, support vector machines can be considered an advantageous estimation method since they can generalize more details. On the other hand, the artificial neural network algorithm is one of the most commonly used machine learning algorithms in terms of classification, optimization, estimation, regression, and pattern tracking. From this point of view, this study aims to predict greenhouse gas emissions using artificial neural network algorithms and support vector machines by estimating CO2, CH4, N2O, and F-gases from greenhouse gases. The data set was obtained from the Turkish Statistical Institute and the years are included between 1990 and 2019. All analyzes were performed using MATLAB version 2019b software.Publication Metadata only Applying social networks to engineering education(WILEY, 2018) DOĞAN, BUKET; Dogan, Buket; Demir, Onder; Ulku, Eyup E.Social networking sites (SNSs) are a popular Internet-based means for users to communicate and interact with each other. Although they have caught the attention of many researchers and are already being used as educational tools, very few studies have investigated the effects of using an SNS in engineering education. This study, therefore, aims to analyze the effects of using the Edmodo platform as a teaching and learning support tool on students' academic and practical performance in the Introduction to Information Technology and Algorithms course, as well as in the Computer Programming course they took in the following semester. It also considers the students' opinions about the Edmodo system. For this study, a total of 62 students studying in the Electrical and Electronics Engineering Department during the 2016-2017 fall semester were divided into two equally sized groups. The control group underwent a traditional face-to-face education, whereas the experimental group augmented this using the Edmodo system. A mixed-methods approach with a post-test-only control group design was used: quantitative data were obtained from student tests, together with qualitative data from follow-up interviews. The students' grades were analyzed using Student's t-test and correlation analysis, showing that the experimental group performed better in their academic and laboratory assessments and that there was a moderately positive relationship between the post-test results and performance in the subsequent Computer Programming course.Publication Open Access NFT-based Asset Management System(2022-01-01) ÜLKÜ, EYÜP EMRE; Abaci I., ÜLKÜ E. E.There are billions of houses, businesses, and lands in the world, and we can prove the ownership of these assets with title deeds prepared by government offices. In the purchase and sale transactions of these titled assets, it is necessary to go through long and complex possess, and the actions that need to be taken do not end here. The asset must also be insured and paid regularly for insurance, tax, and some subscriptions like electricity, water, natural gas, etc. This study aims to create a blockchain-based asset management system that uses NFTs (non-fungible tokens), smart contracts, and the Ethereum network.Publication Metadata only Detection of suspicious activities on windows systems with log analysis(2022-12-21) ÜLKÜ, EYÜP EMRE; Öztürk A., ÜLKÜ E. E.In recent years, rapid technological developments in many different fields have brought along various problems along with many innovations. One of these problems is cyber-attacks. Storing many records and data in digital media has made it very important to protect these records and data. Continuous log records play an important role in taking necessary precautions against cyber-attacks by system administrators. With the logging mechanism found in Windows systems, every transaction made on the system is recorded. These log records are analyzed with various algorithms and tools. As a result of these analyzes, suspicious or attacker behaviors on the system are detected. In this study, various cyber-attacks were tested in an environment where these Windows systems are located. As a result of these tests, the logs formed in the systems were collected and analyzed with the ELK Stack toolkit. As a result of these analyzes, the attacks were determined and associated with the tactics and techniques on Mitre ATT & CK.