IPB University Logo

SCIENTIFIC REPOSITORY

IPB University Scientific Repository collects, disseminates, and provides persistent and reliable access to the research and scholarship of faculty, staff, and students at IPB University

AI Repository
 
Building and Categories


      View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Mathematics
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Mathematics
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Model Prediksi Hotspot Berbasis Variabel Iklim di Sumatera Selatan dan Sekitarnya Menggunakan XGBoost dengan Fitur Autoregresif

      Thumbnail
      View/Open
      Cover (350.4Kb)
      Fulltext (1.177Mb)
      Lampiran (435.8Kb)
      Date
      2026
      Author
      Rizqi, Gilang Syahrul
      Nurdiati, Sri
      Najib, Mohamad Khoirun
      Metadata
      Show full item record
      Abstract
      Kebakaran hutan dan lahan di Provinsi Sumatera Selatan ditandai oleh kemunculan hotspot yang dipengaruhi kondisi iklim dan pola waktu. Penelitian ini bertujuan menganalisis pengaruh variabel iklim terhadap jumlah hotspot serta membangun model prediksi menggunakan Extreme Gradient Boosting (XGBoost) dengan fitur autoregresif. Data deret waktu bulanan periode 2001-2025 diolah melalui prapemrosesan, seleksi fitur, dan validasi berbasis waktu. Hasil menunjukkan bahwa kelembapan tanah dan curah hujan berkorelasi negatif terhadap jumlah hotspot, sedangkan hari tanpa hujan dan indeks iklim global berkorelasi positif. Model XGBoost dengan fitur autoregresif memberikan kinerja terbaik dengan Explained Variance Score sebesar 78,9% dan kesalahan prediksi lebih rendah dibandingkan model tanpa fitur autoregresif. Hasil ini menunjukkan bahwa fitur autoregresif dapat meningkatkan akurasi prediksi hotspot.
       
      Forest and land fires in South Sumatra are indicated by hotspot occurrences influenced by climate conditions and temporal patterns. This study aims to analyze the influence of climate variables on hotspot counts and to develop a prediction model using Extreme Gradient Boosting (XGBoost) with autoregressive features. Monthly time series data from 2001 to 2025 were processed through preprocessing, feature selection, and time-based validation. The results show that soil moisture and rainfall are negatively correlated with hotspot counts, while dry days and global climate indices are positively correlated. The XGBoost model with autoregressive features achieved the best performance with an Explained Variance Score of 78.9% and lower prediction errors than the model without autoregressive features. These findings indicate that autoregressive features improve hotspot prediction accuracy.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/173395
      Collections
      • UT - Mathematics [105]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository