Download Yuan - GeoSpatial and GeoTemporal Informatics

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Pattern recognition wikipedia , lookup

Data (Star Trek) wikipedia , lookup

Agent-based model in biology wikipedia , lookup

Time series wikipedia , lookup

Agent-based model wikipedia , lookup

Cognitive model wikipedia , lookup

Transcript
GeoSpatial and GeoTemporal Informatics
for dynamic and complex systems
May Yuan
What is solved
• Integration of spatial and temporal data
– Location and snapshot
– Dimensional, typological, and typological issues
• Empirical and computational solutions
–
–
–
–
–
–
Spatiotemporal statistics / time geography
Spatial data mining
Machine learning and artificial intelligence
Cellular automata
Agent-based modeling
Evolutionary algorithms
What is almost solved?
• Refined observation networks and data assimilation
– Smart sensor networks
– Adaptive sensor networks
– Data-model assimilation
• Refined statistical and modeling approach to
account for geographic dynamics and complexity
–
–
–
–
Data stream mining
Intelligent agents
Geographically aware agents
Adaptive agents
What has failed? Include surprised failures
•
•
•
•
Fail is a strong word. Perhaps, appears limited
Space-time composite model
Intelligent highways
Integration of sensor networks across scales (local,
regional, and global) to automatically relate
phenomena and processes across scales
• Communication among heterogeneous networks
– Climate
– Ecology
– hydrology
What is missing (discuss areas not
currently on the radar)
• From space-time observations to spatiotemporal processes
• Informatics framework to automate recognition of events and
processes
• Analysis of events and processes
– Weather (obs > events > systems > severity)
– Population (obs > migration > gentrification)
trigger
event
process
drive
state
spatiotemporal data
measured by
sensors
What next? Include both high risk and
needed topics
• P2P sensor networks
• Informatics Theory of Geographic Dynamics and Complexity
• Ontology
• Representation: flows, vector fields
• Analytical and computational approaches that address multi-scalar
relationships among observations, geographic dynamics, and
geographic systems
• Modeling approaches to address interactions among dynamics of
different kinds across scales and domains
– Climate and ecology
– Politics and sociology
– Climate and society
– New levels of knowledge representation
– Aggregates, agglomerates, and narratives
Narratives and sensing making
• By telling and listening to stories, we
communicate, learn, and make sense of the world
by ordering events and assimilating them to find
meanings and implications (Mateas and Sengers
2003).
• Narrative GIS that can tell stories
• Visual means to tell stories