Pratama, Dio Riandi (2024) Analisa Prediksi Titik Api Dengan Menggunakan Metode Deep Learning Di Pulau Sumatra. Other thesis, Universitas Islam Riau.
![[thumbnail of 183510020.pdf]](https://repository.uir.ac.id/style/images/fileicons/text.png "183510020.pdf") |
Text
183510020.pdf - Submitted Version Restricted to Registered users only Download (2MB) | Request a copy |
Abstract
Indonesia, a tropical country, frequently experiences forest fires, especially on the islands of Sumatra and Kalimantan, which have negative impacts on the environment and people's lives, including decreasing air quality and damage to ecosystems. Most fires are caused by human activities such as deforestation and land burning. This research focuses on the island of Sumatra, using historical data to map hotspots and analyzing them separately from data for Indonesia as a whole. Using the LSTM method in Deep Learning and NASA datasets from 2017-2023 processed with Python 3.8 in Jupyter IDE, this research aims to produce a prediction model for hotspots in 2024 on Sumatra Island to assist in prevention strategies and provide valuable information to the community and related parties other. The results of predicting hotspot data in Sumatra using the LSTM method with 80% training data and 20% testing data showed an MSE error rate of 10.43% which means the prediction accuracy level reached more than 89%
| Item Type: | Thesis (Other) |
|---|---|
| Contributors: | Contribution Contributors NIDN/NIDK Sponsor Evizal, Evizal 1029027601 |
| Uncontrolled Keywords: | Forest fires, Sumatra Island, Fire hotspots, NASA, Long Short Term Memory (LSTM). |
| Subjects: | T Technology > T Technology (General) |
| Divisions: | > Teknik Informatika |
| Depositing User: | Furqan nafis al-azami |
| Date Deposited: | 09 Sep 2025 04:03 |
| Last Modified: | 09 Sep 2025 04:03 |
| URI: | https://repository.uir.ac.id/id/eprint/28000 |
Actions (login required)
 |
View Item |