The best practice for handling geospatial data is putting them into a spatially enabled database like PostGIS. RapidMiner can handle database accesses well.
There is also a GeoProcessing extension that can read Shapefiles and calculate distances, areas, intersections etc. However, if you already have ArcGIS and/or a database, you would probably keep your geographic processing there.
Yes, image classification is possible in RapidMiner, but there is no specialization on spatial images.
I just tried to read a shapefile with RM 9.10 and the extension and it worked. You might need to check an option in the settings: Settings => Preferences => Start-Up => Grant additional permissions to unsigned extensions
Answers
It seems that ArcGIS has an Rest API endpoint so you will be able to pull data from it through the web mining extension with POST requests.
For the image mining you'll need to download the Image Handling extension + some deep learning.
Maybe @BalazsBarany could guide us further.
There is also a GeoProcessing extension that can read Shapefiles and calculate distances, areas, intersections etc. However, if you already have ArcGIS and/or a database, you would probably keep your geographic processing there.
Yes, image classification is possible in RapidMiner, but there is no specialization on spatial images.
I just tried to read a shapefile with RM 9.10 and the extension and it worked.
You might need to check an option in the settings:
For image classification you will probably use the Image Handling extension together with Deep Learning.
https://marketplace.rapidminer.com/UpdateServer/faces/product_details.xhtml?productId=rmx_image_handling
https://marketplace.rapidminer.com/UpdateServer/faces/product_details.xhtml?productId=rmx_deeplearning
Check the documentation in Studio and for Deep Learning the videos in the Academy.
Regards,
Balázs