You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Aug 25, 2022. It is now read-only.
I have a simple service running on 10.5, collecting all tweets within the rough extent of Australia and pushing to the spatiotemporal ArcGIS Data Store.
Issue: I am getting a large number of tweets recorded at a single location within the major cities.
For example in Melbourne:
This is not the center of Melbourne, it is in fact a residential street. And the points are all in the exact XY location. From inspecting some of the underlying tweets using the id_str (example), I can see that there is a pattern. They have all opted to specifying a placename.
This is the same in all of the other major Cities. It leads to a very large skew in the data, as this is not the actual location of the tweet, but a general placename.
Looking through the Twitter Doc, users now have the option to either 'Share precise location' which would use the Location Services (GPS) from the device, or they can just tag a generic location - ' attach a location (such as a city or neighborhood) of your choice to your Tweet'.
I would assume that the Connector itself would filter out these tweets that are not a precise location?
I can't see any incoming property on the tweet in GeoEvent that would let me filter these out. There are too many of these generic locations to filter them all out using tiny spatial filters.
I have a simple service running on 10.5, collecting all tweets within the rough extent of Australia and pushing to the spatiotemporal ArcGIS Data Store.
Issue: I am getting a large number of tweets recorded at a single location within the major cities.
For example in Melbourne:
This is not the center of Melbourne, it is in fact a residential street. And the points are all in the exact XY location. From inspecting some of the underlying tweets using the id_str (example), I can see that there is a pattern. They have all opted to specifying a placename.
This is the same in all of the other major Cities. It leads to a very large skew in the data, as this is not the actual location of the tweet, but a general placename.
Looking through the Twitter Doc, users now have the option to either 'Share precise location' which would use the Location Services (GPS) from the device, or they can just tag a generic location - ' attach a location (such as a city or neighborhood) of your choice to your Tweet'.
I would assume that the Connector itself would filter out these tweets that are not a precise location?
I can't see any incoming property on the tweet in GeoEvent that would let me filter these out. There are too many of these generic locations to filter them all out using tiny spatial filters.
I have recorded a screencast here which includes audio. (Apologies, Screencast.com requires Flash)
The text was updated successfully, but these errors were encountered: