Penerapan Machine Learning Untuk Klasifikasi Tingkat Kematangan Buah Anggur (Vitis) Dengan Metode K-Nearest Neighbor

Abstract

Grapes are fruit plants in the form of vines belonging to the Vitaceae family that can be eaten directly or processed into drinks and food. Grapes have color characteristics to determine the level of ripeness of fruit correctly. Determination of ripeness of grapes is usually done directly by looking at the color of ripeness of the fruit grapes that can be seen from the color red if it is ripe and green if it is not ripe. And this could be a mistake in classifying and determining the level of fruit maturity. Classification is a way of grouping objects based on the characteristics possessed by the object of classification. Classification is carried out by computers using the methods used to classify. K-nearest neighbor is a method that can be used in classification and is a machine learning algorithm with a supervised learning approach that works by classifying new data using the similarity between new data and a number of data (k) at the closest available location. KNN algorithm is used for classification and regression. By using the majority category, the classification results from the three best data can be labeled as targets according to the initial dataset. The best data is data to 2, 3, 10 of these data, there are 1 category of immature and 2 ripe, so that the majority of the classification results are ripe grapes. So the classification results for test image data 11 are grapes included in the ripe classification category.