Focus

Validazione delle comunità nei processi di mappatura AI assistita. Il progetto pilota fAIr di OSM e Microsoft in Uganda, Tanzania, Kenya e Nigeria

Parte del Focus "Nuove tecnologie e Intelligenza artificiale. Sfide per la pianificazione"
a cura di Adriano Bisello e Michele Grimaldi

Chiara Centanaro

Assegnista di ricerca, DAD/Università degli Studi di Genova

DOI

https://doi.org/10.62661/ui317-2024-045

Keywords

Adattamento, Intelligenza artificiale, Partecipazione comunitaria

Abstract

L’inclusione delle comunità nella mappatura e implementazione di tecnologie open-source per generare dati affidabili nei piani di mitigazione aprono a nuovi processi di validazione. La Volunteered Geographic Information e l’Intelligenza artificiale stanno trasformando la gestione dei territori, soprattutto in contesti privi di cartografie ufficiali. OpenStreetMap e Humanitarian OpenStreetMap permettono la raccolta e validazione collaborativa di dati spaziali, favorendo l’inclusione delle comunità locali. Il progetto fAIr introduce un modello open-source di mappatura AI assistita in Kenya, Nigeria e Tanzania per migliorare la resilienza urbana. FAIr utilizza i feedback delle comunità per migliorare le mappe generate con strumenti open-source di machine learning, affrontando bias nei modelli di apprendimento e potenziando le capacità decisionali delle comunità. L’integrazione di questi strumenti nei piani urbanistici consente di identificare aree a rischio e sviluppare strategie di mitigazione.

Riferimenti bibliografici

Centanaro C., (2023), “Real-time, crowd-sourced online maps in disaster management in De-sign Envi-ronment Landscape City 2021”, in S. Eriche, G. Pellegri (eds.), Venice Biennale Resilient Communities Conference Proceedings, Aracne Editrice, Roma, p. 175-186.
Coleman D., Georgiadou Y., Labonte J. (2009), “Volunteered geographic information: The nature and mo-tivation of produsers”, International Journal of Spatial Data Infrastructures Research, vol. 4, p. 332-358.
Crawford S., Goldsmith S. (2014), The Responsive City – Engaging Communities Through Data-Smart Governance, Jossey-Bass, San Francisco.
Dar Ramani Huria (2016), Ramani Huria Flood Resilience Atlas for Dar es salaam [https://documents1.worldbank.org/cu...].
Gausa M., Guallart, V., Muller, W., Soriano F., Porras, F. et al. (2003), The Metapolis dictionary of advanced architecture: City, Technology and Society in the Information Age, Actar, Barcelona.
Gevaert C., Persello C., Nex F., Vosselman G. (2018), “A deep learning approach to DTM extraction from imagery using rule-based training labels”, ISPRS Journal of Photogrammetry and Remote Sensing, no. 142, p. 106-123. http://doi.org/10.1016/j.isprsjprs.2018.06.001
GFDRR Labs - Global Facility for Disaster Reduction and Recovery (2020), “Open Cities AI Challenge Dataset, Version 1.0”, Radiant MLHub. https://doi.org/10.34911/rdnt.f94cxb
Goodchild M. F. (2007), “Citizens as sensors: the world of volunteered geography”, GeoJournal, no. 69, p. 211-221. https://doi.org/10.1007/s10708-007-9111-y
Goodchild M. F. (2009), “NeoGeography and the nature of geographic expertise”, Journal of Location Based Services, vol. 3, no. 2, p. 82-96. https://doi.org/10.1080/17489720902950374
Heeks R., Shekhar S. (2019), “Datafication, development and marginalised urban communities: an ap-plied data justice framework”, Information, Communication & Society, vol. 22, no. 7., p. 992-1011. https://doi.org/10.1080/1369118X.2019.1599039
Holderness T. (2014), “Geosocial Intelligence”, Technology and Society Magazine, IEEE, no. 33, p. 17-18. http://dx.doi.org/10.1109/MTS.2014.2301860
HOT - Humanitarian OpenStreetMap Team (2022), Satellite imagery for social good in Kenya and Nigeria [https://www.hotosm.org/projects/sat...].
HOT - Humanitarian OpenStreetMap Team (2024), Satellite Imagery For Social Good Nigeria [https://www.youtube.com/watch?v=Zy_...].
Minghini M., Liu P., Li H., Grinberger A.Y., Juhász L. (eds) (2022), “State of The Map”, in Proceedings of the Academic Track, Florida International University, Florence, Italy, 19-21 august 2021 [https://zenodo.org/records/7004791].
Nassozi S. (2022), Satellite Imagery for Social Good - Our Reflections in State of the map 2022 [https://media.ccc.de/v/sotm2022-186...].
OSM - OpenStreetMap (2024), Tags [https://wiki.openstreetmap.org/wiki/Tags].
Raj Sharma K. (2023), fAIr - Free and Open Source AI for Humanitarian Mapping [https://www.youtube.com/watch?v=j8q...].
TURP - Tanzania Urban Resilience Program (2019a), The Msimbazi Opportunity Transforming the Msim-bazi Basin into a Beacon of Urban Resilience. Volume B. Detailed Plan for the Lower Basin, World Bank Group, Washington [https://documents1.worldbank.org/cu...].
TURP - Tanzania Urban Resilience Program (2019b), The Msimbazi Opportunity Transforming the Msim-bazi Basin into a Beacon of Urban Resilience. Public Disclosure for the Green Lung of Dar es Salaam. Volume C Appendices, World Bank Group, Washington [https://documents1.worldbank.org/cu...].
Zanchetta A. (2024), “Assessing the performance of AI-assisted mapping of building footprints for OSM”, State of the Map 2024 [https://2024.stateofthemap.org/sess...].