Subscribe to DSC Newsletter

Hello everyone,

I am currently starting some research on human behavior modeling and prediction. While searching for the best statistics and data mining software I came across a very big $$ issue :) As I am doing this in the course of my PhD and currently the institute/university is not capable of providing a license for it, I decided to go for R.

I am particularly enthusiastic as it can be plugged together with Java and therefore address stuff in real-time. Based on your expertise, do you think this software will limit my results? What you be the major drawbacks of R?

Best Regards,
Jose Simoes

Views: 806

Reply to This

Replies to This Discussion

Just a small update.

after doing some small research I found out that R has 2 plugins that allow interconnecting with Java, they are:

rJava - http://www.rforge.net/rJava
Rserve - http://www.rforge.net/Rserve

They allow both using R to invoke java code and vice-versa (what I was looking for).

In the following weeks I will try to run some code in a Java Application Server and post the results here.

Once again, thank you all for helping me out.

Jose
Here comes the promised update!

Although with a bit delay I managed to implement the scenario I mentioned earlier. I used the Rserve libraries on both client and server instance.

Basically I am running some Java code running on a Sailfin application server and I can make calls to R using this code. In this way, I developed some web service interfaces that can be called in a fancy Flash application.

To sum up, the scenario is tested (in terms of technology) and will further evaluate weather it works good on loaded environments and real-time systems.

Stay tuned!
Jose

PS: If anyone needs help setting up a similar environment, don't hesitate to contact me.
One of the major drawbacks of R is the learning curve. If anyone is finding that curve too steep, I'd recommend the R Inferno book as it explains many of the languages peculiarities (particularly the unhelpful error messages).

I've combined R with a Ruby on Rails application. The obstacles I've found are:
- the need to fork the rails process so that the server can return messages to the browser without having to wait for RScript to complete
- intensive disk usage by R slowing the rest of the server (ionice and nice - for cpu rationing - help)
- passing parameters to R and receiving errors from it, I'm using a clumsy combination of argv[] (parameter type is determined by sequence like $1 $2 $3 in bash) and stdout/ stderr and would welcome suggestions of an alternative method.

I've not felt the need to use RServe as my application is designed to provide a non-technical front end to some R scripts rather an online stats/analytical engine.
Hi Jose,

We understand & recognise your resource constraints, It should be responsibilities/duties of institution/unvesrties to facilitate/invest on the requiered/desired resources by the researcher.

It is not specific to R (Free Ware Software) but most of the available freeware Software, they go through really many of rigorous questionere.

i.e What is the reliability/accuracy of the result ?
What are the product development life cycle they follow ?
What are the benchmarking they qualify.
How many research paper have been published.
It has been gone through proper beta/pilot testing so on & so forth. Which discourage/limit the researcher to apply freeware software for academecally approved research work.

If we may be any of your assistance, please let us know.

Thanks & Wish you all the best.
Biswajit.

RSS

On Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service