Can I publish RapidMiner results with Educational Edition license?

Casper72

I'm a PhD student (Health Science), currently working on predictive analytics based on patient reported outcome measures. My application for a RapidMiner Studio Educational Edition license has been granted, and so far I really like it! My question is: can I publish results obtained with RapidMiner, without violating the license agreement? By publishing I mean submitting articles to journals. My thesis is not sponsored by any commercial corporations, and I do not intend to monetize on any results.

varunm1

Hello @Casper72

I work on rapidminer for most of my ML related works and published some papers as well. The only thing you need to do is to cite rapidminer in your references and where relevant. You can also go through our discussion about citing rapidminer below.

https://community.rapidminer.com/discussion/54779/how-to-cite-rapidminer-studio-for-scholarly-publications

Hope this helps

Varun

SGolbert

Hi @Casper72

there is no problem with publishing, but there could be problems on not publishing the processes/source code (which is in any case bad for any research, especially in the health area). Note that RM has a lot of AGPL extensions, so you have to be careful if you use RM as part of some bigger program or workflow.

Also I have seen some papers that "publish" a screenshot of the RM process, that's a very bad practice. Just include the process file or at least the XML code.

Regards,

Sebastian

DocMusher

Casper, FYI

Do not hesitate to contact me if I can help you with your health science project and RM. Good luck with your PhD!

https://journals.plos.org/plosone/article/authors?id=10.1371%2Fjournal.pone.0145791

Cheers Sven

Casper72

Thank you very much varunm1!

Casper72

Thank you Sebastian, that is a very good point.

Casper72

Hi Sven, 

Much appreciated! Just read your article and it is very informative  Currently I have finished developing and validating a total of 45 logistic regression and MARS-models - 5 outcome measures x 9 diagnoses. I would like to see how well they stand up to more advanced methods (Machine Learning). For that purpose I think RapidMiner's Auto Model feature would be an excellent and time saving starting point. 

Casper.