Perbandingan Kinerja Pengklasifikasian KNN Dan SVM Pada Dataset Citra Rosa

SANAM, Agatha Yanri (2024) Perbandingan Kinerja Pengklasifikasian KNN Dan SVM Pada Dataset Citra Rosa. Undergraduate thesis, Universitas Katolik Widya Mandira Kupang.

	Text ABSTRAK.pdf Download (856kB)
	Text BAB I.pdf Download (687kB)
	Text BAB II.pdf Restricted to Repository staff only Download (547kB)
	Text BAB III.pdf Restricted to Repository staff only Download (732kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (852kB)
	Text BAB V.pdf Restricted to Repository staff only Download (616kB)
	Text BAB VI.pdf Download (184kB)
	Text DAFTAR PUSTAKA.pdf Download (497kB)

Abstract

Roses or scientifically called rosa are a type of plant that has thorny stems and layered petals. Its shape is round or oval. In this research, rosa images were classified using the K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) methods. The aim of the research is to compare the classifier performance of KNN and SVM on rosa images dataset. The dataset used is 1000 images with a division of training data and testing data with a ratio of 80%:20%. This research uses two tests, namely 3-fold cross validation and random sampling. The results of the 3-fold cross validation test for the KNN method are Accuracy 99,8%, Precision 99,5%, Recall 99,5%, and F1- Score 99,5%, meanwhile, the test results for the SVM method are Accuracy 99,2%, Precision 98%, Recall 98%, and F1-Score 98%. The 3-fold cross validation tests were successfully carried out with the Accuracy, Precision, Recall and F1-Score values on KNN being superior to the Accuracy, Precision, Recall and F1-Score values on SVM.

Actions (login required)

View Item