Integrating Multi-Variable Driving Factors to Improve Land Use & Land Cover Classification Accuracy using Machine Learning Approaches - A Case Study from Lombok Island

Aug 5, 2026

Accurate land cover classification underpins effective land management and environmental monitoring. This study improves land cover classification for Lombok Island by applying advanced machine learning models, including Random Forest, Gradient Boosting, Decision Tree, and Naive Bayes, while integrating Landsat satellite imagery, spectral indices, physiographic, climatic, and socioeconomic variables. Random Forest achieved the highest model accuracy at 82% and the best field validation overall accuracy at 88%. Enhanced multi-variable modeling also helps reduce satellite cloud-cover issues, supporting sustainable planning and conservation decisions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Introduction

Methods

Study area

Software

Variables

Jurnal Manajemen Hutan Tropika, 31(2), 123-132, May 2025 EISSN: 2089-2063 Scientific Article ISSN: 2087-0469 DOI: 10. 7226/jtfm.31.2. 123 Integrating Multi-Variable Driving Factors to Improve Land Use & Land Cover Classification Accuracy using Machine Learning Approaches: A Case Study from Lombok Island Miftahul Irsyadi Purnama1,3*, Hüseyin Oğuz Çoban2 1Department of Forest Engineering, The Institute of Graduate Education, Isparta University of Applied Sciences, Isparta, Türkiye 32260 2Department of Forest Engineering, Faculty of Forestry, Isparta University of Applied Sciences, Isparta, Türkiye 32260 3Saujana Climate Community, West Lombok, Indonesia 83363 Received September 3, 2024/Accepted March 12, 2025 Abstract Accurate classification of land cover is essentialfor effective landmanagement and environmental monitoring. This study aimed to enhance land cover classification for Lombok Island using advanced machine learning algorithms. The models employed include Random Forest, Gradient Boosting, Decision Tree, and Naive Bayes, integrating a wide range of variables, such as Landsat satellite imagery, spectral indices, physiographic, climatic, and socioeconomic data. Among these, Random Forest demonstrated the highest model accuracy at 82%, followed by Gradient Boosting at 80%, Decision Tree at 73%, and Naïve Bayes at 61%. In field validation assessments, comparing thepredictions of these machine learning models with ground truth data, Random Forest was the most reliable, achieving an overall accuracy of 88%. This superior performance is largely due to the multi-variable approach, which allows the model to mitigate issues like cloud cover in satellite images. The key variables that significantly influenced the land cover classification on Lombok Island include proximity to settlements, temperature, and distance to roads. These results provide essential insights for land management strategies, enabling policymakers and stakeholders to make informed decisions on sustainable development, urban planning, and environmental conservation in rapidly changing landscapes. Keywords: land cover classification, machine learning,Random Forest, LombokIsland *Correspondence author, [email](email: miftahulpurnama@gmail.com)[:](email: miftahulpurnama@gmail.com)[ miftahulpurnama@gmail.com](email: miftahulpurnama@gmail.com) Introduction The rapid advancements in machine learning (ML) technologies have revolutionized various fields, including environmental science and land management. Accurate land cover classification is essential for effective environmental monitoring, sustainable land use planning, and conservation efforts (Vinaykumar et al., 2023) . However, achieving high accuracy in land cover classification remains challenging due to the complex and dynamic nature of landscapes (Desjardins et al. , 2023) . Traditional methods often struggle to accommodate the variability in land cover types, leading to inaccuracies that can significantly impact decision-making processes (Gavade & Rajpurohit, 2021; Qichi et al., 2023) . Lombok Island, a region experiencing rapid urbanization, agricultural expansion, and environmental change, presents a unique case for studying land cover dynamics (Rahayu et al., 2023) . The island's diverse ecosystems and the pressures from human activities require a robust and accurate classification system to manage and protect its natural resources effectively (Dewi & Sukmawati, 2020) . The integration of ML approaches with multi-variable driving factors, such as climate data, topography, and socio- economic variables, offers a promising solution to improve the precision of land cover classification (Jaya et al., 2015; Mitra & Basu, 2023) . ML is favored over traditional methods because it automates data analysis, efficiently handling vast variablesand generating new insights. Unlike classical techniques, which struggle with complex datasets, ML excels in processing and predicting

FAQs

Which machine learning models were used to classify land cover on Lombok Island?

The study used Random Forest, Gradient Boosting, Decision Tree, and Naive Bayes. These models were trained with multi-variable inputs to improve classification performance.

How did Random Forest perform compared with the other models?

Random Forest achieved the highest model accuracy at 82% and the best field validation overall accuracy at 88%. Gradient Boosting followed with 80% model accuracy, while Decision Tree and Naive Bayes were lower.

What factors and data were most important for the classification results?

The study integrated Landsat imagery, spectral indices, physiographic, climatic, and socioeconomic data. Key influencing variables included proximity to settlements, temperature, and distance to roads, and the multi-variable approach helped mitigate issues such as cloud cover.

Aug 5, 2026

Upload by River Wang