Jurnal Informatika dan Teknologi Komputer ( J-ICOM)
Vol 4 No 1 (2023): Jurnal Informatika dan Teknologi Komputer ( JICOM)

Penerapan Backpropagation Neural Network pada Prediksi Curah Hujan di Sumatera Utara

Ulya Nabila (Universitas Samudra)
Indah Ramadhani (Universitas Samudra)
Fazrina Saumi (Universitas Samudra)



Article Info

Publish Date
01 Jul 2024

Abstract

Indonesia is located on the equator with a latitude of 6°N to 11°S. This latitude causes a high intensity of solar radiation so that the air temperature becomes high. This causes the evaporation of water from the surface of the sea and land to form clouds and rain. One of the provinces in Indonesia that has high rainfall is North Sumatera. Due to the high rainfall, the province often experiences floods and landslides. In order to minimize the impact of floods and landslides, the first thing that can be done is to predict the rainfall in North Sumatera. The method used is Backpropagation Neural Network. This method is a kind of artificial intelligence designed to process information by imitating the nervous system of the human brain. This research aimed to predict rainfall in North Sumatera using Backpropagation Neural Network and determine the accuracy of this method. The data came from the Badan Pusat Statistik (BPS) of North Sumatera, namely rainfall data from January 2020 to December 2022. The results implied that rainfall in 2023 for each month is 190; 266; 197; 178; 290; 182; 299; 350; 213; 485; 357; and 389 (mm/month) and has an accuracy of Mean Absolute Percentage Error (MAPE) of 35.55%, so this method is categorized as suitable for predicting rainfall.

Copyrights © 2023






Journal Info

Abbrev

jicom

Publisher

Subject

Description

Jurnal J-ICOM (Jurnal Informatika dan Teknologi Komputer) dimaksudkan sebagai media kajian ilmiah hasil penelitian, pemikiran dan kajian analisis-kritis mengenai penelitian Rekayasa Sistem, Teknik Informatika/Teknologi Informasi, Manajemen Informatika dan Sistem Informasi. Sebagai bagian dari ...