Yazar "Adem, Kemal" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • [ X ]
    Öğe
    A Comparative Analysis of Deep Learning Parameters for Enhanced Detection of Yellow Rust in Wheat
    (Kırıkkale Üniversitesi, 2024) Adem, Kemal; Kavalcı Yılmaz, Esra; Ölmez, Fatih; Çelik, Kübra; Bakır, Halit
    Wheat, one of the most important food sources in human history, is one of the most important cereal crops produced and consumed in our country. However, if diseases such as yellowpas, which is one of the risk factors in wheat production, cannot be detected in a timely and accurate manner, situations such as decreased production may be encountered. For this reason, it is more advantageous to use decision support systems based on deep learning in the detection and classification of diseases in agricultural products instead of experts who perform the processes in a longer time and have a higher error rate. In this study, the effects of the number of layers, activation function and optimization algorithm variables on the classification of deep learning models used for the classification of yellow rust disease in wheat were examined. As a result of the study, the highest success value was obtained with 97.36% accuracy when using a 5-layer CNN model using Leaky ReLU activation function and Nadam optimization algorithm.
  • [ X ]
    Öğe
    Diagnosis of Chronic Kidney Disease using Random Subspace Method with Particle Swarm Optimization
    (Kırıkkale Üniversitesi, 2018) Adem, Kemal
    Late diagnosis of chronic kidney disease, adisease that has increased in recent years and threatens human life, may leadto dialysis or kidney failure. In this study, kNN, SVM, RBF and Random subspacedata mining methods were applied on the data set consisting of 400 samples and24 attributes taken from UCI for classification of chronic kidney disease with particleswarm optimization (PSO) based feature selection method. As a result of thestudy, the results of the application of each data mining method are comparedwith the resultant training and test results. As a result of the comparison, itwas seen that the method of PSO feature selection affects the classificationsuccess positively. Moreover, as a method of data mining, it has been seen thatthe random subspace method has higher accuracy rates than the other methods.
  • [ X ]
    Öğe
    Faster R-CNN Kullanarak Elmalarda Çürük Tespiti
    (Kırıkkale Üniversitesi, 2019) Cömert, Onur; Hekim, Mahmut; Adem, Kemal
    Bu çalışmada, elmalardan alınan görüntülerüzerinde evrişimsel sinir ağı yöntemlerinden olan Faster R-CNN kullanılarakelmaların çürük ve sağlam olarak sınıflandırılması amaçlanmaktadır. Önerilenmodelde işlem adımları sırasıyla görüntü alma-önişleme, çürük bölgelerin tespitedilmesi ve elmaların sınıflandırması şeklindedir. Görüntü alma-önişlemeaşamasında, tasarlanan bir görüntü alma platformu içerisinde bulunan NIR kamerakullanılmıştır. Çalışmada 100’ü çürük ve 100’ü sağlam olan toplam 200 adetelmanın her birinin 6 farklı açısından toplam 1200 adet görüntü eldeedilmiştir. Önişleme aşamasında, bu görüntülere sırasıyla uyarlamalı histogrameşitleme, kenar bulma, morfolojik işlemler uygulanmıştır. Önişlem uygulanarakgörünürlüğü iyileştirilen yeni görüntüler kullanılarak eğitilen Faster R-CNNmodeli ile çürük bölgeler tespit edilmiştir. Sınıflandırma aşamasında, çürük vesağlam elmaların tespit edilmesinde %84,95 doğru sınıflandırma oranınaulaşılmıştır. Sonuç olarak, önerilen modelin meyve suyu gıda sanayisinde çürükve sağlam elmaların otomatik olarak tespit edilmesinde kullanılabileceğidüşünülmektedir. 
  • [ X ]
    Öğe
    Sentiment Analysis of Comments on Courses on Massive Online Course Platforms using Text Mining
    (Kırıkkale Üniversitesi, 2023) Daşgın, Ramazan; Adem, Kemal
    In our age, things that are easily accessible, less costly, and have no time and space restrictions attract more attention. Considering the possibilities that people have recently and the development of technology, this change in people's interest has also been reflected in education. People now want to access content that they can choose from wherever they want, whenever they want. As a result of these requests, Massive Open Online Course (MOOC) platforms began to emerge. There are many paid or free courses on these platforms. Before enrolling in these courses, many people register based on the comments made and the score given to the course. However, it is not easy to decide about a course by reading all the reviews. In this study, comments made on Turkish courses on Udemy, one of the MOOC platforms, were used in order to evaluate the courses positively and negatively without the need for users to read the comments. On these comments, positive and negative evaluations were made about the courses using classical machine learning and deep learning. With BayesNet, J48 and OneR algorithms from classical machine learning, the most successful result was obtained from BayesNet algorithm with an accuracy of 91.576%. After applying Random, GloVe and Word2Vec word embeddings to the dataset, hybrid architectures of GRU and CNN-LSTM from deep learning models were applied and the most successful result was obtained from GRU architecture with an accuracy of 95.67% after using GloVe word embedding.