PENERAPAN K-NEAREST NEIGHBORS PADA PERCEPTRON UNTUK KLASIFIKASI DATASET KECIL DENGAN TIGA FITUR

Arjuna Ranma Putra
Julya Rahmah Shanty
Maria Lodiana Leki
Muhammad Aqshal Hidayat Fizhillillah
Rifky Firmansyah
Devi Yunita

OKTAL : Jurnal Ilmu Komputer dan Sains
Vol 3 No 11 (2024): OKTAL : Jurnal Ilmu Komputer Dan Sains

In developing machine learning models for small datasets, choosing the right method is key to producing accurate classification. This research applies the Single Layer Perceptron (SLP) algorithm to classify a small dataset with three main features, namely Feature1, Feature2, and Feature3. The SLP algorithm is used to learn patterns in the data, with model evaluation using the k-fold cross-validation technique. This technique ensures each piece of data is used as test and training data in turn, to obtain more accurate evaluation results. In addition, the k-Nearest Neighbor (k-NN) algorithm was used to find the optimal K parameter value to improve the accuracy of the model. This study used 13 sample data to train and test the model.