RapidMiner

Native statistics operators

SOLVED
Super Contributor

Native statistics operators

Hello,

 

I was curious - is there a free set of native operators for rapidminer (not the paid Old World Computing extensions) that are more about statistics such as generating descriptives, frequencies, covariance, correlations, ANOVA, parametric, non-parametric tests, etc.? I can do all this in my free copy of excel provided by my university but I'd like to have these statistic results alongside the modeling operators in my process in rapidminer.

 

Thank you.

9 REPLIES
Moderator

Re: Native statistics operators

[ Edited ]
I would check out the marketplace to see if there are any free extensions. As far as I know the statistics extension from old world computing is the only one. It does fill a gap that the native RapidMiner platform doesn't have.

You could always reach out to old world computing and ask if they have an educational license.
RM Certified Expert

Re: Native statistics operators

Another option would be to use the R scripting extension (which is free) and use R to do any specific statistical testing that you want.  That would have no license cost, although it is a bit more cumbersome than using the Old Word Computing extension (which is excellent, by the way).

 

Brian T., Lindon Ventures - www.lindonventures.com
Analytics Consulting by Certified RapidMiner Analysts
Highlighted
Community Manager

Re: Native statistics operators

I will chime in only to say that Sebastian's (Old World Computing) stats extension is very good.  It saves me a lot of time when I need to do descriptive stats like deciles and such.  Well worth the $$.

 

Scott

Scott Genzer
Senior Community Manager
RapidMiner, Inc.
Super Contributor

Re: Native statistics operators

Thank you Scott but as a broke grad student, I find it not worth it to spend that kind of money for software. I'm extremely thankful to the RapidMiner team for making it free for students and I hope that developers in and around the product will uphold that value proposition of Rapidminer when making their own extensions. I like the Rapidminer model of being a quasi open-source software - paid for corporations and free for academic use.

Super Contributor

Re: Native statistics operators

Thank you Thomas for your suggestion. Let's say I got a group of people together and we developed our own version of the said statistics extensions and we wanted to make them available for free in Rapidminer. What would it take to do so such that you guys maintain the code thereafter for the next RM versions and not me or my friends?

Moderator
Solution
Accepted by topic author batstache611
‎05-18-2017 02:48 PM

Re: Native statistics operators

You could build it and put it on a webpage or put it on github. This way people can edit it as the RM versions change. 

 

 

Community Manager

Re: Native statistics operators

@batstache611 yes I think we all can empathize with the broke grad student situation.  When I was a grad student, I used to buy a large fried rice from the local Chinese takeout place and eat it for a week because that was all I could afford.  Smiley Happy

 

It would certainly not hurt to reach out to Sebastian at Old World Computing and see what he can do for folks in your situation.  As we all know, poor grad students often (hopefully) become successful business people who will be controlling budgets later in their careers.  And yes we are all huge supporters of RapidMiner's academic licensing for exactly this purpose.

 

Good luck with your work.

 

Scott

Scott Genzer
Senior Community Manager
RapidMiner, Inc.
Moderator

Re: Native statistics operators

I echo @sgenzer's thoughts. Sebastian is a pretty cool guy and was once a very broke Uni student. I would reach out to him and ask if there's a temp educational liscense he can grant you. 

Super Contributor

Re: Native statistics operators

Thank you for your encouragement @sgenzer and @Thomas_Ott, I will reach out to Sebastian and ask if he can help me out. Thank you and I'm sorry if I didn't exactly sound polite there in my previous response.