Analisis Metode K-Nearest Neighbor Menggunakan Rapid Miner Untuk Memprediksi Hujan Kota Surakarta

Alvian Ahmada Akhbar (Universitas Duta Bangsa Surakarta)
Dwi Hartanti (Universitas Duta Bangsa Surakarta)

Prosiding Seminar Nasional Teknologi Informasi dan Bisnis
Prosiding Seminar Nasional Teknologi Informasi dan Bisnis (SENATIB) 2023

This study aims to implement the K-Nearest Neighbor (KNN) algorithm using the Rapid Miner application to predict rain in Surakarta City. Rainfall has an erratic pattern making it difficult to predict manually. Rainfall cannot be determined with certainty but this can be estimated. Thus, the existence of Data Mining can enable machines to recognize and study complex data patterns. Therefore program learning can learn patterns of rainfall data to make some predictions. This study uses three variables as criteria, namely temperature, wind speed, and humidity. The test results using the K-Nearest Neigbor (KNN) algorithm and the Rapid Miner application with a value of K = 3, found an accuracy of 83.87%. From 31 data taken in July 2023. The results of the analysis prove that the KNN method using the Rapid Miner application can be used to predict rain in Surakarta City.