Open hsu000001 opened 9 years ago
@amarnath will explore with @hsu000001 offline. Capabilities exercised: iterative search, benefit of enhancers and ontologies to value added search (more complete discovery). Address data heterogeneity (labeling of 'bedload tracer experiment')
bedload-tracer experiment: tag clasts in a channel, and see where they go over some period of time. Trying to measure how fast/far material is carried by the river; typically associated with variou steam flow events. There is not alot of this data, so generalize 'bedload tracer experiment' in original language to 'sediment transport experiment/observation' [2015-03-04]
may not have data sources for bedload tracer data; there should be CZO datasets we could harvest to run this. “at that location”-- CZO site, vicinity of site. need to infer location of experiment (one location per dataset? or multiple experiments in datasets)
Use Case: CZO Community: Find data from all bedload tracer experiments in the field, and any other studies done at those locations.
Analysis
Text parse:
‘Find’ => provide a list of resource descriptions ‘data’ == observation result ‘experiment’ == observation ‘data from ...experiments’ => resouce of interest is observation results ‘bedload tracer’ == phenomenon of interest (what is wanted is not actually clear--this requires background knowledge to resolve) ‘in the field’ => sampling feature is a naturally occurring feature (not a laboratory feature like a flume) ‘study’ => resource (an accessible representation of some research) ‘studies..at..locations’ => resources with a target extent that overlaps location of any ‘Find’ result. This might be a sampling feature extent or a feature of interest extent.
Queries
Find observation datasets with [phenomenon IsA bedload tracer](this require some concept expansion) AND [sampling feature IsA natural feature] For each dataset select observations where [phenomenon IsA bedload tracer] For each observation get sampled feature location For each location find resources where [resource extent Overlaps the location] Detail: if resource a dataset, select features or observations in the dataset where [feature of interest OR sampling feature extent Overlaps the location]