English Türkçe

Quick Access


Bu Dergi DOI ve Crosscheck üyesidir


Elma Çeşitlerinin Sınıflandırılması: H2O Tabanlı Kollektif Öğrenme ve Naive Bayes Algoritmalarının Karşılaştırılması
(Classification of Apple Varieties: Comparison of Ensemble Learning and Naive Bayes Algorithms in H2O Framework )

Author : Abdullah BEYAZ  Dilara GERDAN, Abdullah BEYAZ, Mustafa VATANDAŞ  
Type :
Printing Year : 2020
Number : 2020-1
Page : 9-16


Summary
Bu çalışmada, H2O tabanlı makine öğrenmesi sınıflandırma teknikleri, meyveleri kabuk rengine göre sınıflandırmak amacıyla kullanılmıştır. Veri seti oluşturmak için rastgele seçilen 60 adet Red Delicious, 60 adet Golden Delicious ve 60 adet Granny Smith elma çeşidine ait veriler değerlendirilmiştir. Meyve renk değerlerinde, L *, a * ve b * renk uzayı esas alınmış ve taşınabilir spektrofotometre ile ölçümler yapılmıştır. Veri analizi için H2O Gradyan Artırma Makinesi, H2O Rastgele Orman ile H2O Naive Bayes algoritmaları seçilmiştir. Veri seti test için% 30 eğitim için ise % 70 olarak bölümlendirilmiştir. Değerlendirme; doğruluk (%), yüzde hata (%), F-Ölçümü, Cohen's Kappa, hatırlama, doğruluk, doğru pozitif (TP), yanlış pozitif (FP), gerçek negatif (TN), yanlış negatif (FN) değerleri gibi performans göstergelerine göre yapılmıştır. Sonuçlar, H2O Gradyan Artırma Makinesi için % 100,0, H2O Rastgele Orman için % 98,4 ve H2O Naive Bayes algoritması için % 100,0 doğrulukta elde edilmiştir.

Turkish Keywords
Elma sınıflama, H2O makine öğrenmesi, Gradyan Artırma Makinesi, Rastgele Orman, Naive Bayes

Abstract
In this study, H2O machine learning classification techniques were used to classify the apples according to skin color of the fruits. For each variety 60 samples were used at evaluations of the fruits. Fruit color values were based on L *, a * and b * color space and measured by a portable spectrophotometer. Randomly, Red Delicious, Golden Delicious and Granny Smith apple varieties were studied to create the database. H2O Gradient Boosting Machine, H2O Random Forest and H2O Naive Bayes Algorithms were used to for data analysis. The data set was partitioned to 30% for testing and 70% for training. The classifier performance which accuracy (%), error percentage (%), F-Measure, Cohen’s Kappa, recall, precision, true positive (TP), false positive (FP), true negative (TN), false negative (FN) values were given at the conclusion section of the research. The results found that 100,0 % accuracy for H2O Gradient Boosting Machine, 98,4 % accuracy for H2O Random Forest and 100,0 % accuracy for H2O Naive Bayes.

Keywords
Apple classification, H2O machine learning, Gradient Boosting Machine, Random Forest, Naive Bayes

Advanced Search


Announcements















    [JAFAG] Yeni Sayımız Basılmıştır JAFAG, 36 (1)

    [JAFAG] New Issue Published JAFAG, 36 (1)



    Call for papers

    To Researcher:

    I am delighted to invite you to disseminate your research findings in Journal of Agricultural Faculty of Gaziosmanpasa University (JAFAG).  I wish to publish original research articles in agricultural sciences. In our journal: rapid manuscript publication and peer review process with online manuscript submission system, quick access of your research results to global audience  with open access system.

    I hope you will contribute your article for publication in JAFAG and thanking you in advance.


    AIMS and SCOPE

    Journal of Agricultural Faculty is scientific, peer reviewed journal and belonged to the Gaziosmanpasa University Faculty of Agriculture since 1985. Journal of Agricultural Faculty of Gaziosmanpasa University aims to publish the national and international scientific paper on agriculture sciences (agricultural economics, animal science, biosystems engineering, field crops, fisheries engineering, horticulture, plant protection, soil science and plant nutrition). The journal is published the third issues in a year from 2014.


    Journal of Agricultural Faculty of Gaziosmanpasa University (JAFAG)

    Journal of Agricultural Faculty of Gaziosmanpasa University (JAFAG) is an international peer-reviewed journal and belonged to the Gaziosmanpasa University Faculty of Agriculture since 1985.

    The journal is open access.

    The journal has both online and printed publishing.

    The journal has a manuscript submisson system.

    There are no page charges.


    INDEXING

    ACADEMIC JOURNALS DATABASE

    AGRICOLA

    ARASTIRMAX

    CAB ABSTRACTS

    CROSSREF

    DIRECTORY OF SCIENCE

    DOAJ (DIRECTORY OF OPEN ACCESS JOURNALS)

    DRJI (DIRECTORY OF RESEARCH JOURNAL INDEXING)

    EBSCO

    EFITA (INTERNATIONAL DIRECTORY OF AGRICULTURE, FOOD AND THE ENVIRONMENT)

    ELECTRONIC JOURNALS LIBRARY

    FAO- AGRIS

    GENAMICS JOURNALSEEK

    GLOBAL IMPACT AND QUALITY FACTOR

    GOOGLE SCHOLAR

    INDEX COPERNICUS

    JOURNAL INDEX

    JOUR INFORMATICS

    OPEN ACCESS LIBRARY

    OPEN J-GATE

    RESEARCH BIBLE

    ROAD (DIRECTORY OF OPEN ACCESS SCHOLARLY  RESOURCES)

    SIS (SCIENTIFIC INDEXING SERVICE)

    SCIENCE CENTRAL

    THE GROVE ONLINE LIBRARY

    TUBITAK-ULAKBIM

    ULRICHSWEB

    WORLDCAT


    [JAFAG] New Issue Published JAFAG, 36 (1)
    Journal of Agricultural Faculty of Gaziosmanpasa University, JAFAG has just published its latest issue (Vol 36, No 1, 2019, JAFAG) at http://ziraatdergi.gop.edu.tr.

    ICV Impact Factor: 4.95 (2012) - 5.94 (2013) - 70.06 (2015) - 72.50(2016) - 84.65 (2017)

    We invite you to review the Contents here and then visit journal web site to review articles.

    Thanks for the continuing interest in our work in journal.



Address :Gaziosmanpaşa Üniversitesi Ziraat Fakültesi Dergisi Yayın Ünitesi, Tokat-TÜRKİYE / Adress:Journal of Agricultural Faculty of Gaziosmanpasa University Publishing Unit, Tokat, TURKEY
Telephone :03562521616-2178 / Telephone: +90 356 252 16 16- 2 Fax :03562521488 / Fax:+90 356 252 14 88
Email :ziraatdergi@gop.edu.tr/ e-mail: ziraatdergi@gop.edu.tr

Web Yazılım & Programlama Han Yazılım Bilişim Hizmetleri