22

I've used the integration of R with QGIS, and found it a very uesful tool. Having taken a look at the R toolbox extension for 9 and the extension for 10 it appears they both offer only point clustering to demonstrate the proof of concept.

Can other analysis methods available in R be implemented in ArcGIS Desktop through a python script (for example, implementing specific parts of the spatstat package)?

PolyGeo
  • 65,136
  • 29
  • 109
  • 338
om_henners
  • 15,642
  • 2
  • 46
  • 86

3 Answers3

8

There is now an R-ArcGIS Community on Github that offers:

  • r-bridge-install to:

Install the R ArcGIS Tools

  • r-bridge:

Bridge library to connect ArcGIS and R, including arcgisbinding R library

  • r-sample-tools:

Sample tools illustrating R usage in geoprocessing scripts

PolyGeo
  • 65,136
  • 29
  • 109
  • 338
7

The Geospatial Modelling Environment (the successor of Hawth's Tools for ArcGIS) links python scripting, R and ArcGIS in a useful way. I haven't investigated the links with R in detail, but it looks like it may be useful for what you're trying to do.

If you're trying to do something yourself then StatCONN may be useful.

robintw
  • 4,006
  • 11
  • 40
  • 60
3

I extended Mark Janikas' work on interfacing R and ArcGIS.

Specifically, I built an "R script to ArcGIS Toolbox converter." The user writes an annotated R script and loads it to an ArcMap add-in, which generates a ToolBox for the R script. The ToolBox can be used to load results to ArcMap.

This is the work flow:

enter image description here

And this is how the ArcMap add-in looks like:

enter image description here

enter image description here

Compared to the Mark Janikas' work, my approach avoids two steps for R users wanting to share their work:

  • Defining a toolbox from scratch in ArcToolbox (where each parameter has to be set by hand)
  • Creating a Python script with the logic of the created toolbox, so the user doesn't have to know anything about Python.

Look at this video to get a better idea.

Germán Carrillo
  • 36,307
  • 5
  • 123
  • 178
  • 1
    You know that Mark's scripts are depreciated, correct? The logistic function calls two depreciated functions and the cluster tool relies on a package that no is no longer maintained and as such, is not available at R >= 3.0.0. Given that R is at 3.1.2, users will likely get nothing but errors when running through your example. I have rewritten these scripts with several notable improvements including an autologistic model and silhouette optimization using a K-medoids clustering algorithm. If you would like the scripts, let me know and I will aim you to the download. – Jeffrey Evans Dec 15 '14 at 18:54
  • I took Mark Janikas' script just for the sake of comparing both approaches, his and mine, through a common example. I expect users use their own R scripts (they would need to annotate them first) to experiment and take advantage of my add-in. Thanks anyway for your offer. – Germán Carrillo Dec 15 '14 at 20:23