#geoai — Public Fediverse posts

Using Geospatial Analysis and Explainable Machine Learning to Examine Risk Factors of Out-of-Hospital Cardiac Arrest Survival Outcome
--
https://openscholar.uga.edu/record/26998/files/Zhang_uga_0077E_16145.pdf <-- shared technical publication / dissertation
--
https://www.aag.org/award-grant/william-l-garrison-award-for-best-dissertation-in-computational-geography/ <-- shared @AAG William L. Garrison Award for Best Dissertation in Computational Geography
--
[again, way outside any expertise I might have, but fascinating spatial analysis use case…]
H/T @Jielu Zhang | Postdoctoral Researcher @ Harvard University
“[The authors] Ph.D. dissertation "Using Geospatial Analysis and Explainable Machine Learning to Examine Risk Factors of Out-of-Hospital Cardiac Arrest Survival Outcome" [1st link above] has received the 2026 biennial William L. Garrison Award for Best Dissertation in Computational Geography from the American Association of Geographers… [2nd link above.]
In [their] research, [they] develop[ed] Explainable GeoAI and Causal GeoAI methods that combine geographic data and artificial intelligence to expose and ameliorate health disparities by delivering models that not only predict risks but also illuminate how and where to intervene. While [their] dissertation focused on cardiovascular disease, these approaches are broadly applicable to public health, medicine, urban planning, environmental exposure, and resilience research…”
#explainable #causal #AI #model #modeling #PublicHealth #GIS #spatial #mapping #spatialanalysis #spatiotemporal #AAG2026 #AAG #Award #geostatistics #Georgia #health #risk #hazard #cardiacarrest #cardiovacscular #usecase #metrics #midocine #urbanplanning #resilience #survival #OutofHospital #PhD #Dissertation #CardiacArrest #AutomatedExternalDefibrillator #SpatialOptimization #GeographicallyExplainableArtificialIntelligence #GeoAI #GeoXAI #SpatiallyAwareCausalInference #OverlayedSpatioTemporalOptimization #healthcare #medical #intervention #GIS #spatial #mappingt #spatialanalysis #spatiotemporal #heart #heartattack #AED #survival #survivaloutcomes #machinelearning #AI #publichealth #healthgeographers #counterfactual #explainable #deeplearning #model #modeling

The Myth And Reality Of ‘Physics-Based’ Models In Geosciences
--
https://geoaiunpacked.substack.com/p/the-myth-and-reality-of-physics-based <-- shared technical article
--
“Across hydrology, fire, weather, and climate science, models are often described as physics-based—a phrase that evokes rigor, transparency, and trust. Yet, beneath the surface, much of what drives these models are empirical relationships and parameterizations rather than direct solutions of physical laws. These assumptions are necessary to make the equations computable, but they also introduce biases, tuning artifacts, and uncertainties that can propagate through every layer of simulation.
This doesn’t mean physics-based models are flawed or obsolete; rather, it highlights the complexity of modeling natural systems…”
#GeoAI #datadriven #physicsbased #model #modeling #usecase #appliedscience #accuracy #appropriateness #simulation #naturalsystems #hybrid #physicalrealism #hydrology #fire #weather #climate

The quest for a fair TimeGPT benchmark

At the end of yesterday's #TimeGPT for mobility post, we concluded that TimeGPT's trainingset probably included a copy of the popular BikeNYC timeseries dataset and that, therefore, we were not looking at a fair comparison ...

http://anitagraser.com/2025/03/29/the-quest-for-a-fair-timegpt-benchmark/

#GeoAI #MobilityDataScience #FoundationModels #GISChat

#TimeGPT for #mobility: Can foundation models outperform classic machine learning models for flow predictions?

tldr; Maybe. Preliminary results certainly are impressive.

http://anitagraser.com/2025/03/28/can-foundation-models-outperform-classic-machine-learning-models-for-mobility-predictions/

#MobilityDataScience #gischat #urbanmobility #geoai #foundationmodels

New read 📝: Thinking #Geographically about #AI #Sustainability

"We propose a framework to evaluate
models from several sustainability-related angles, including #energyefficiency, carbon intensity, #transparency, and social implications. We encourage future AI / #GeoAI work
to acknowledge its #environmental impact as a step towards a more resource-conscious society.

https://agile-giss.copernicus.org/articles/4/42/2023/agile-giss-4-42-2023.pdf

#agile2023 #gischat #ethics #SpatialDataScience #geography

GeoAI In The US Geological Survey For Topographic Mapping
--
https://doi.org/10.1111/tgis.12830 <-- shared paper
--
https://www.sciencebase.gov/catalog/item/5ec6aad082ce476925eddbc6 <-- download training dataset
--
#GIS #spatial #mapping #GeoAI #GeoNat #AI #ML #artficialintelligence #machinelearning #machinelearningalgorithms #data #training #research #artificialintelligence #development #geospatial #gischat #topomap #dataextraction #hydrology #water #terrain #maps
@USGS