Implementation of Supervised Learning Method In Grapevine Leaf Classification

Abstract

Grapevine leaves are a type of leaf variety that is difficult to identify because it will take time if processed manually so research will be carried out using the help of machine learning. This research aims to classify 5 varieties of grapevine leaves using orange data mining and several classification methods namely k-Nearest Neighbors (kNN), logistic regression, random forest and support vector machine (SVM). The dataset used is 500 images and 5 classes where each class consists of 100 images, namely Ak (100), Buzgulu (100), Ala_Idris (100), Dimnit (100), and Nazli (100). The stages in the analysis process are to enter the image into orange data mining by passing several stages so that the image dataset can be processed and read on the test and score so that the confusion matrix can be obtained. The results of the research conducted using orange data mining show that classification using logistic regression gives the best results at a precision value of 0.848% and a recall value of 0.847%. This research shows that classification using orange data mining also provides good results, besides that this research can also help in the classification process so that it does not require a long time.