Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Aulia Herdhyanti
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Prediksi Curah Hujan dengan Empat Parameter menggunakan Backpropagation (Studi Kasus: Stasiun Meteorologi Ahmad Yani) Aulia Herdhyanti; Lailil Muflikhah; Imam Cholissodin
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 12 (2022): Desember 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Rainfall is the amount of water that falls to the surface of the ground over a certain period. The rainfall itself is recorded by the BMKG (Meteorology, Climatology and Geophysics Agency) every day. According to research, rainfall is influenced by several parameters, including the temperature, the humidity of the air, the speed of the wind, and the activity of solar. In forecasting the weather, high accuracy is needed because the weather greatly affects the activities of the population. High rainfall can cause floods. Because of this problem, one solution that can solve this problem is to predict the rainfall with the backpropagation method which is one of the neural network architectures that has a multi-layer network and processes the training data forward and corrects the errors backwards. This study uses the rainfall data with parameters that influence it, namely the average temperature, the humidity of the air, the speed of the wind, and the activity of solar within 19 months from the Ahmad Yani Meteorological Station in Semarang. The best accuracy obtained with the backpropagation method is the MSE value of 0.006952 which was obtained by using 2 hidden neurons, the maximum iteration is 1000 iterations, the amount of training data is 70% of the total dataset, and the learning rate is 0.05.
Co-Authors Imam Cholissodin Lailil Muflikhah