Person: YILDIZ, KAZIM
Loading...
Email Address
Birth Date
Research Projects
Organizational Units
Job Title
Last Name
YILDIZ
First Name
KAZIM
Name
13 results
Search Results
Now showing 1 - 10 of 13
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 Open Access Comparison of the machine learning methods to predict wildfire areas(2022-09-01) YILDIZ, KAZIM; BAYAT G., YILDIZ K.In the last decades, global warming has changed the temperature. It caused an increasing the wildfire in everywhere. Wildfires affect people's social lives, animal lives, and countries' economies. Therefore, new prevention and control mechanisms are required for forest fires. Artificial intelligence and neural networks(NN) have been benefited from in the management of forest fires since the 1990s. Since that time, machine learning (ML) methods have been used in environmental science in various subjects. This study aims to present a performance comparison of ML algorithms applied to predict burned area size. In this paper, different ML algorithms were used to forecast fire size based on various characteristics such as temperature, wind, humidity and precipitation, using records of 512 wildfires that took place in a national park in Northern Portugal. These algorithms are Multilayer perceptron(MLP), Linear regression, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree and Stacking methods. All algorithms have been implemented on the WEKA environment. The results showed that the SVM method has the best predictive ability among all models according to the Mean Absolute Error (MAE) metric.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 Metadata only Detection of DDoS attacks with feed forward based deep neural network model(PERGAMON-ELSEVIER SCIENCE LTD, 2021) BULDU, ALİ; Cil, Abdullah Emir; Yildiz, Kazim; Buldu, AliAs a result of the increase in the services provided over the internet, it is seen that the network infrastructure is more exposed to cyber attacks. The most widely used of these attacks are Distributed Denial of Service (DDoS) attacks that easily disrupt services. The most important factor in the fight against DDoS attacks is the early detection and separation of network traffic. In this study, it is suggested to use the deep neural network (DNN) as a deep learning model that detects DDoS attacks on the sample of packets captured from network traffic. DNN model can work quickly and with high accuracy even in small samples because it contains feature extraction and classification processes in its structure and has layers that update itself as it is trained. As a result of the experiments carried out on the CICDDoS2019 dataset containing the current DDoS attack types created in 2019, it was observed that the attacks on network traffic were detected with 99.99% success and the attack types were classified with an accuracy rate of 94.57%. The high accuracy values obtained show that the deep learning model can be used effectively in combating DDoS attacks.Publication Metadata only Comparison of LDA, NMF and BERTopic topic modeling techniques on Amazon product review dataset: A case study(2023-07-21) YILDIZ, KAZIM; BÜYÜKTANIR, BÜŞRA; TURAN S., YILDIZ K., BÜYÜKTANIR B.Publication Open Access Identification of wool and mohair fibres with texture feature extraction and deep learning(INST ENGINEERING TECHNOLOGY-IET, 2020-02) YILDIZ, KAZIM; Yildiz, KazimWool and mohair fibres are both animal-based fibres and having circular scales on their microscopic images from the longitudinal view. Although they look very similar in their microscopic view, they show different physical/chemical properties which determine their usage area. Thus, in textile industry, they need to be separated carefully from each other. The separation of wool/mohair fibres is an important issue and can be performed with human eye by using the microscopic images, that is not time/cost effective and not objective. The novelty of the presented study is to design an objective, easy, rapid, time and cost-effective method in order to separate wool fibre from mohair fibre by using a texture analysis based identification method. For this purpose, microscopic images of both wool and mohair fibres were preprocessed as the texture images. Local binary pattern-based feature extraction process and deep learning were separately used to get determinative information from the fibres. In order to identify the samples, the classification based method was completed. Experimental results indicated that an accurate texture analysis for this kind of animal fibres is possible to identify wool and mohair fibres by using deep learning and machine learning with 99.8% and 90.25% accuracy rates, respectively.Publication Open Access Detection of autistic spectrum disorder using artificial neural network(2023-08-01) YILDIZ, KAZIM; ÖZDEMİR Ş. N., YILDIZ K.Autistic Spectrum Disorder (ASD) is a neuro-developmental disorder that is congenital or manifests with a delay in social relations and physiological development at an early age, and also causes problems in communication. It is possible to reduce the effect of the disease on individuals with early diagnosis. However, detecting ASD at an early age requires time and cost. In the studies conducted in recent years, it is seen that there is a serious increase in ASD cases. In order to prevent this increase, decision support systems should be established for early diagnosis. It is important to develop decision support models to diagnose ASD, especially for children aged 12-36 months. In this study, a model was developed that can help in detecting ASD with high accuracy for 12-36 months old children. The data set used in the created model was collected from the mobile application named ASDTests developed by Thabtah. In the estimation phase, four different machine learning algorithms which are support vector machine, Naive Bayes,Random Forest and Artificial Neural Network were used. In the classification process, high success rate was obtained with artificial neural network, random forest classifier.Publication Open Access Powdery mildew detection in hazelnut with deep learning(2022-09-01) YILDIZ, KAZIM; BOYAR T., YILDIZ K.Hazelnut cultivation is widely practised in our country. One of the major problems in hazelnut cultivation is powdery mildew disease on hazelnut leaves. In this study, the early detection of powdery mildew disease with the YOLO model based on machine learning was tested on a unique data set. Object detection on the image, which is widely applied in the detection of plant diseases, has been applied for the detection of powdery mildew diseases. According to the results obtained, it has been seen that powdery mildew disease can be detected on the image. Using YOLOv5, diseased areas were detected with approximately 90% accuracy in diseased leaf images. Multiple leaves in one image were detected with approximately 85% accuracy in detecting healthy areas using images with complex backgrounds. The model, which has been used in different studies for the detection of disease in plant leaves, also gave effective results in the detection of powdery mildew disease in hazelnut leaves. Early detection of powdery mildew with a method based on machine learning will stop the possible spread of disease. It will increase the efficiency of hazelnut production by preventing the damage of hazelnut producers.Publication Open Access Evaluation of nano-filler dispersion quality in polymeric films with binary feature characteristics and fractal analysis(INST ENGINEERING TECHNOLOGY-IET, 2020-08) YILDIZ, KAZIM; Yildiz, Kazim; Yildiz, ZehraThis study investigates the use of binary features and fractal dimension analysis to evaluate the dispersion quality of nanofillers in thin polymeric films by using light microscopy images. For this purpose, polymeric films were cast with the inclusion of various montmorillonite (MMT) nanofiller amounts. Then the light microscopy images were captured from the polymeric films then preprocessed for the evaluation. Thresholding process was applied to the obtained images for each nanofiller percentage level. The obtained binary level images were used in the feature extraction process with binary statistics and fractal dimension. Thermogravimetric analysis (TGA) was used to evaluate the flame resist behaviour of polymeric films based on the dispersion quality of nanofillers. The samples with various nanofiller contents were tested using the image processing method and the results were all compared with the TGA results. The results obtained by the feature extraction process and TGA, about the dispersion quality of nanofillers, were all in good agreement.Publication Open Access Housing price estimation with deep learning: A case study of Sakarya Turkey(2022-06-01) YILDIZ, KAZIM; BÜYÜKTANIR, BÜŞRA; ÖZDEMİR M., YILDIZ K., BÜYÜKTANIR B.Shelter is one of the most basic human needs. Besides housing needs, the housing market is also very important for investment. It is also a market where many people, such as engineers, architects, real estate agents make economic gain. When a house is bought for living in it, it is not desired to be changed for many years, and when it is bought for investment, it is a tool that requires good income. Therefore, the best decision should be made when buying a house, and it should be scrutinized. Correct estimation of house prices is very important for both buyers to make the right decision and for sellers to sell without a loss. There are many parameters for estimating house prices. In addition to variables such as the number of floors, location, and several bathrooms used in previous studies, economic factors (such as the price of bread, foreign currency price, new car price) and the housing loan interest rate of the banks were taken as inputs in this study. Sakarya province, where all parameters can be tested to make a more accurate determination, was chosen as the research area. A comparison of polynomial regression, random forest, and deep learning methods was made and it was concluded that the most accurate method was deep learning. At the same time, it was determined which parameters are more effective in house price estimation.