Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal G-Tech : Jurnal Teknologi Terapan
Yunifa Miftachul Arif
Universitas Islam Negeri Maulana Malik Ibrahim Malang, Indonesia
Computer Science & IT Decision Sciences, Operations Research & Management Energy Engineering

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

Found 1 Documents
Search

 1 
Rainfall Classification in Malang Regency Using Artificial Neural Networks with Boolean Logic-Based Feature Engineering Selina Ayuningtyas; Zainal Abidin; Yunifa Miftachul Arif
G-Tech: Jurnal Teknologi Terapan Vol 10 No 3 (2026): G-Tech, Vol. 10 No. 3 July 2026
Publisher : Universitas Islam Raden Rahmat, Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70609/g-tech.v10i3.10263

Abstract

This study classifies monthly rainfall in Malang Regency using an Artificial Neural Network (ANN) with the Backpropagation algorithm and Boolean logic approaches (AND, OR, and AND-OR). The dataset consists of 144 monthly climatological records (2012–2023) obtained from the East Java Climatology Station, with three input variables: rainfall, minimum temperature, and relative humidity. Rainfall was grouped into three categories: Low (0–100 mm; 39.58%), Moderate (101–300 mm; 37.50%), and High (301–500 mm; 22.92%). Boolean logic features were generated using the mean values of relative humidity (77.34%) and minimum temperature (17.89°C). The ANN model was tested with three hidden-layer configurations containing 5, 8, and 10 neurons. Data were divided into 70% training and 30% testing sets using a random state of 42. The results show that the 10-neuron configuration achieved the best performance, with 75.00% accuracy, 75.97% precision, 75.00% recall, and 75.00% F1-score. In comparison, the 5-neuron and 8-neuron models achieved accuracies of 68.18% and 65.91%, respectively. The AND-OR Boolean logic approach provided more stable feature representation than the AND or OR approaches alone by combining multiple atmospheric conditions. These findings indicate that ANN with an appropriate architecture can effectively classify rainfall patterns in Malang Regency.
Co-Authors Selina Ayuningtyas Zainal Abidin