Analysis K-Nearest Neighbor Method in Classification of Vegetable Quality Based on Color

Abstract

In this research, the process of applying the K-Nearst Neighbor (KNN) method will be carried out, which is a classification method for a collection of data based on the majority of categories and the goal is to classify new objects based on attributes and sample samples from training data. So that the desired output target is close to the accuracy in conducting learning testing. The results of the test of the K-Nearest Neighbor method. It can be seen that from the K values ??of 1 to 10, the percentage of the results of the analysis of the K-NN method is higher than the results of the analysis of the K-NN method. And from the K value that has been tested, the K 2 value and the K 9 value have the largest percentage so that the accuracy is also more precise. As for the results of testing the K-Nearest Neighbor method in data classification. As for the author's test using a variation of the K value of K-Nearest Neighbor 3,4,5,6,7,8,9. Has a very good percentage of accuracy compared to only K-NN. The test results show the K-Nearest Neighbor method in data classification has a good percentage accuracy when using random data. The percentage of variation in the value of K K-Nearest Neighbor 3,4,5,6,7,8,9 has a percentage of 100%.