At the New York Technology Summit in May 2013, I presented a different way of integrating QlikView with R leveraging extensions and other open source technologies such as Open CPU. This new way will allow for R functions to be centrally managed via a server (Open CPU) and be easily deployed and called by the end users by using our ajax zero footprint browser. Basically, QlikView will call R on demand, whenever a user makes a selection within the dashboard. My research builds upon the previous example posted by Elif Tutuk, "Integrating QV with R example kit.zip" (http://community.qlik.com/docs/DOC-2975), the example utilizes the statconnDCOM Noncommercial API as well as Brian Bauer, "QlikView and R Integration for Predictive Analytics Example" (http://community.qlik.com/docs/DOC-3981).
My approach consist of 3 different systems integrating to each other.
1. OpenCPU: is an initiative from Jeroen Ooms and has been implemented by Stat/Dev consulting in The Netherlands. The implementation has been supported by a Research and Development grant (WBSO) from the Dutch government. OpenCPU sits ontop of R and makes R a restful web service where any web technology can do ajax call to R. The benefits of OpenCPU is two folds: a. its free b. you can use their hosted solution in the cloud (basically OpenCPU wil host your R code) or you can download the software and install it on your own ubuntu server (behind your firewall).
2. The R-Project:OpenCPU comes package with R so no work is required to integrate OpenCPU with R
3. An extension within a QlikView application: The final piece to the integration is build a custom extension within your QlikView applicaiton that sends and recieve data to/from open CPU. I have attached a zip file below that contains the following three files:
a. opencpuClusterAnalysis.qar: contains the QlikView extension
b. QlikView-OpenCPU demo.pdf: explains how to get the extension to talk with open CPU (using the hosted solution)
c. QV-OpenCPU Integration Example.qvw: is the QlikView application that needs to be deployed to the QlikView server.
Finally, below is a high level process diagram on how everything comes together.