[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120174-en":3,"doc-seo-120174-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120174,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Uncovering_large_inconsistencies_between_machine_learning_derived_gridded_settlement_datasets - Abstract and findings","High-resolution human settlement maps support disaster response, humanitarian resource allocation, and international development, but precise agreement between machine learning derived gridded settlement datasets remains unclear. This work quantifies overlap for 42 African countries using Open Buildings, High Resolution Population Maps, and GRID3, revealing substantial disagreement on which areas are classified as settled. The study analyzes geographic and socio-economic drivers and trains a model to predict where datasets diverge.","arXiv :2404 . 13 127v 1 [ cs . SI] 19 Apr 2024  \nUncovering large inconsistencies between machine learning derived gridded settlement datasets  \nVedran Sekara 1,* , Andrea Martini2 , Manuel Garcia-Herranz3 , and Do-Hyung Kim3  \n1 Department of Computer Science, IT University of Copenhagen, Copenhagen, Denmark  \n2 United Nations Children’s Fund, Latin America and Caribbean Regional Office, Panama City, Panama  \n3 United Nations Children’s Fund, New York, NY, USA  \n* Corresponding author, [vsek@itu.dk](vsek@itu.dk)  \nABSTRACT  \nHigh-resolution human settlement maps provide detailed delineations of where people live and are vital for scientific and practical purposes, such as rapid disaster response, allocation of humanitarian resources, and international development. The increased availability of high-resolution satellite imagery, combined with powerful techniques from machine learning and artificial intelligence, has spurred the creation of a wealth of settlement datasets. However, the precise agreement and alignment between these datasets is not known. Here we quantify the overlap of high-resolution settlement map for 42 African countries developed by Google (Open Buildings), Meta (High Resolution Population Maps) and GRID3 (Geo-Referenced Infrastructure and Demographic Data for Development) . Across all studied countries we find large disagreement between datasets on how much area is considered settled. We demonstrate that there are considerable geographic and socio-economic factors at play and build a machine learning model to predict for which areas datasets disagree. It it vital to understand the shortcomings of AI derived high-resolution settlement layers as international organizations, governments, and NGOs are already experimenting with incorporating these into programmatic work. As such, we anticipate our work to be a starting point for more critical and detailed analyses of AI derived datasets for humanitarian, planning, policy, and scientific purposes.  \nIntroduction  \nA large range of applications from humanitarian response 1, 2 , urban and infrastructure planning3 , epidemic preparedness4 to environmental science5 , require detailed maps of where people live. However, it can be hard to obtain accurate and timely data6. Traditionally, accurate maps of human settlements have been derived from census data or representative household surveys. Although both sources of information provide rich and irreplaceable data, they can be expensive and time consuming. For instance, the US 2020 national census is estimated to have cost up to $14 .2 billion7. Further, certain areas might not even be surveyable due to e.g. conflicts or natural disasters. Because of these limitations, surveys are only performed once every 5 to 10 or 15 years depending on the country, but rapidly changing contexts, e.g. migration, rapid urbanization, humanitarian crises, and pandemics, can quickly make data outdated.  \nIn response to these limitations developmental agencies, non-governmental organizations (NGOs), academics, and governments have explored new technologies and complementary approaches to estimate population densities8–10. New opportunities have been opened up by increased availability, and better global coverage, of high-resolution satellite imagery, with resolutions down to centimeter level 11, 12. Combined with advances in computation and AI powered image recognition techniques, high-resolution population datasets derived from satellite images have in the recent years become readily available 13, 14. These gridded population estimates provide detailed maps of where people live 15–20. Additionally, some of these datasets also provide insights on the distribution of specific populations, including the number of children under five, the number of women, as well as elderly populations21–23. As such, these population datasets are frequently being used for policy work24 and to monitor the progress of reaching the sustainable development goals","cbCaimwhJZx0HLQp","https://ap.wps.com/l/cbCaimwhJZx0HLQp","pdf",5399630,1,14,"English","en",105,"# Abstract\n# Introduction\n## Need for accurate settlement and population data\n## Limits of census and surveys\n## Satellite imagery and ML-driven datasets\n## Top-down vs bottom-up estimation approaches\n## Role of settlement layers\n# Research approach: quantifying overlap and mismatch","[{\"question\":\"What problem does the document address?\",\"answer\":\"It quantifies how much high-resolution gridded settlement datasets overlap and where they disagree on what areas are considered settled.\"},{\"question\":\"Which datasets are compared for the 42 African countries?\",\"answer\":\"The document compares settlement map layers from Google (Open Buildings), Meta (High Resolution Population Maps), and GRID3.\"},{\"question\":\"What factors influence disagreements between datasets?\",\"answer\":\"It highlights considerable geographic and socio-economic factors, and uses them to build a machine learning model that predicts disagreement regions.\"}]","Uncovering_large_inconsistencies_between_machine_learning_derived_gridded_settlement_datasets - Abstract and findings | PDF",1785728562,35,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"uncovering-large-inconsistencies-between-machine-learning-derived-gridded-settlement-datasets-abstract-and-findings","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/uncovering-large-inconsistencies-between-machine-learning-derived-gridded-settlement-datasets-abstract-and-findings/120174/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the document address?","Question",{"text":75,"@type":76},"It quantifies how much high-resolution gridded settlement datasets overlap and where they disagree on what areas are considered settled.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which datasets are compared for the 42 African countries?",{"text":80,"@type":76},"The document compares settlement map layers from Google (Open Buildings), Meta (High Resolution Population Maps), and GRID3.",{"name":82,"@type":73,"acceptedAnswer":83},"What factors influence disagreements between datasets?",{"text":84,"@type":76},"It highlights considerable geographic and socio-economic factors, and uses them to build a machine learning model that predicts disagreement regions.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]