PENERAPAN ALGORITMA K-NEAREST NEIGHBOR (KNN) DALAM MEMPREDIKSI DAN MENGHITUNG TINGKAT AKURASI DATA CUACA DI INDONESIA

Abstract

Indonesia is a country that has a very large disaster-prone area, so that it can be dubbed the 1001 disaster country. It was recorded that throughout January 2020 there were 518 earthquakes, while in February 2020 there were an increase of 799 earthquakes. It is not just earthquakes that occur in Indonesia, floods, landslides, extreme rainfall, and very drastic temperature changes that occur in Indonesia every year. Hydroclimatologically, Indonesia is also affected by the phenomenon of ENSO (EL_Nino Southern Oscillation) and La Nina, resulting in floods, landslides, drought, and low temperatures (cold). This research will implement the K-NN Algorithm in Predicting and Calculating the Level of Accuracy of Weather Data in Indonesia. The data used is 3623 data which is then divided into training data and testing data with a ratio of 80: 20, 80% for training data or as much as 2898 data from 3623 data and for testing data as much as 20% or as much as 725 data from 3623 data. produce a prediction with a data accuracy rate of 0.8993 or about 89%. With a data accuracy rate of 89%, it is hoped that it can help predict temperature and weather in Indonesia, so that it can help breeders and farmers to reduce the risk of crop failure and disadvantage.