{"title":"Logistic regression versus XGBoost for detecting burned areas using satellite images","authors":"","doi":"10.1007/s10651-023-00590-7","DOIUrl":null,"url":null,"abstract":"<h3>Abstract</h3> <p>Classical statistical methods prove advantageous for small datasets, whereas machine learning algorithms can excel with larger datasets. Our paper challenges this conventional wisdom by addressing a highly significant problem: the identification of burned areas through satellite imagery, that is a clear example of imbalanced data. The methods are illustrated in the North-Central Portugal and the North-West of Spain in October 2017 within a multi-temporal setting of satellite imagery. Daily satellite images are taken from Moderate Resolution Imaging Spectroradiometer (MODIS) products. Our analysis shows that a classical Logistic regression (LR) model competes on par, if not surpasses, a widely employed machine learning algorithm called the extreme gradient boosting algorithm (XGBoost) within this particular domain.</p>","PeriodicalId":50519,"journal":{"name":"Environmental and Ecological Statistics","volume":"19 1","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2024-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Environmental and Ecological Statistics","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1007/s10651-023-00590-7","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Classical statistical methods prove advantageous for small datasets, whereas machine learning algorithms can excel with larger datasets. Our paper challenges this conventional wisdom by addressing a highly significant problem: the identification of burned areas through satellite imagery, that is a clear example of imbalanced data. The methods are illustrated in the North-Central Portugal and the North-West of Spain in October 2017 within a multi-temporal setting of satellite imagery. Daily satellite images are taken from Moderate Resolution Imaging Spectroradiometer (MODIS) products. Our analysis shows that a classical Logistic regression (LR) model competes on par, if not surpasses, a widely employed machine learning algorithm called the extreme gradient boosting algorithm (XGBoost) within this particular domain.
期刊介绍：
Environmental and Ecological Statistics publishes papers on practical applications of statistics and related quantitative methods to environmental science addressing contemporary issues.
Emphasis is on applied mathematical statistics, statistical methodology, and data interpretation and improvement for future use, with a view to advance statistics for environment, ecology and environmental health, and to advance environmental theory and practice using valid statistics.
Besides clarity of exposition, a single most important criterion for publication is the appropriateness of the statistical method to the particular environmental problem. The Journal covers all aspects of the collection, analysis, presentation and interpretation of environmental data for research, policy and regulation. The Journal is cross-disciplinary within the context of contemporary environmental issues and the associated statistical tools, concepts and methods. The Journal broadly covers theory and methods, case studies and applications, environmental change and statistical ecology, environmental health statistics and stochastics, and related areas. Special features include invited discussion papers; research communications; technical notes and consultation corner; mini-reviews; letters to the Editor; news, views and announcements; hardware and software reviews; data management etc.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
