Penerapan Algoritma K-Nearest Neigbors Untuk Klasifikasi Dana Desa
JURNAL INFORMATIKA, Gorontalo 26 November 2016 PENERAPAN ALGORITMA K-NEAREST NEIGHBOR UNTUK KLASIFIKASI DANA DESA Email : Zemibadu@gmail.com ABSTRACT K-Nearest Neighbor algorithm (k-NN) is a method to perform the classification of objects based on the learning data that were located closest to the test data. K-Nearest instance Neighbor including group instance-based learning. The composition of the training data = 80% and testing data = 20% at random, modification K values for classification of parameter changes the accuracy of KNN. Where the results of each changes in the value of k, namely: a) k = 3, accuracy = 78.95%; Based on the evaluation model of k-nearest neighbor using Confusion Matrix, the use of models k-nearest neighbor to the dataset taken that are used in the research object gain accuracy of 78.95% or included Failure by using the parameters k = 2. Then the value of Precision as big as 100% and Recall as big as 100%. Based on these results it can be stated