Enhancing Classification Accuracy of Pumpkin Seed with Detail Morphological Features and Different Machine Learning Algorithms


Coşkun H.

in: NEW FRONTIERS IN ENGINEERING, Akdemir Bayram,Özkaya Umut, Editor, Duvar Yayınları, İzmir, pp.255-274, 2023

  • Publication Type: Book Chapter / Chapter Research Book
  • Publication Date: 2023
  • Publisher: Duvar Yayınları
  • City: İzmir
  • Page Numbers: pp.255-274
  • Editors: Akdemir Bayram,Özkaya Umut, Editor
  • Kütahya Health Sciences University Affiliated: Yes

Abstract

In this study, classification was carried out using morphological features to recognize the types of pumpkin seeds (Koklu et. al., 2021). New morphological feature values were obtained using the open-access pumpkin seed feature data set. Three different data sets were used, using the features in this study and those in the previous study. Gradient boosting, support vector machine, k-nearest neighbors, and random forest machine learning techniques were used to increase classification  success. As a result, models with higher classification metric values than previous study were created.