Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization International Journal Of Science, Technology & Management (IJSTM)
Henry, Amir Acalapati
Unknown Affiliation
Aerospace Engineering Agriculture, Biological Sciences & Forestry Automotive Engineering Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Mechanical Engineering Transportation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : JOIV : International Journal on Informatics Visualization

 1 
Performance Improvement for Hotspot Prediction Model Using SBi-LSTM-XGBoost and SBi-GRU-XGBoost Sukmana, Husni Teja; Aripiyanto, Saepul; Alamsyah, Aryajaya; Henry, Amir Acalapati; Nandaputra, Riandi
JOIV : International Journal on Informatics Visualization Vol 9, No 6 (2025)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.9.6.3047

Abstract

Forest fires damage ecosystems and harm all living beings, often triggered by low rainfall that worsens fire spread. Climatic factors such as the El Nino–Southern Oscillation (ENSO) also contribute to reduced rainfall and prolonged dry seasons. This study aims to enhance the performance of fire prediction models to support forest fire mitigation. Modified artificial neural network algorithms—Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) with bidirectional stacked layers—are employed as baseline models. An experimental approach was used to compare the performance of LSTM and GRU models with their ensemble versions, where XGBoost was added to improve prediction accuracy. The results show that the proposed ensemble algorithms significantly outperform the baseline models in multivariate fire prediction. The SBi-LSTM-XGBoost and SBi-GRU-XGBoost models demonstrated more than a 40% performance improvement compared to the original SBi-LSTM and SBi-GRU models. In multivariate modelling, the ensemble models achieved an R-value of 1.0000, with an average MAE of 0.0007, RMSE of 0.0009, and MAPE of 0.0008. This study also identified limitations of the LSTM and GRU models in processing ENSO data due to their non-linearity and weak correlation with hotspot data. As a contribution, our experiments show that integrating XGBoost into LSTM and GRU models effectively overcomes these limitations, significantly improving hotspot prediction accuracy and supporting better forest fire mitigation strategies.
Co-Authors Alamsyah, Aryajaya Angga Aditya Permana Aripiyanto, Saepul Husni Teja Sukmana Maskurudin, Muhamad Nandaputra, Riandi Pradipta, Allan Sufyan, Ammar Syafiq, Zahra