Talk:Predictive Model Markup Language
A suggestion about the controversial "PMML Tools" section
The "PMML Tools" section has been inserted and deleted several times in recent days. Here is one suggestion: Name the section something else, maybe "Industry Support for PMML", and in the section describe the support that several software packages provide for PMML. I think it is important to material in this section about KNIME, STATISTICA, SAS-Enterprise Miner, and other packages that provide PMML support. One possible citation would be http://journal.r-project.org/2009-1/RJournal_2009-1_Guazzelli+et+al.pdf. Other citations would be needed as well. I think in this context, a description of PMML Converter and Transformations Generator would be appropriate. I think a good citation would be http://kdd11pmml.files.wordpress.com/2011/09/p5_kdd2011_pmmlworkshop_guazzelli_et_al1.pdf. For SAS-EM's PMML support, one possible reference would be Bill Frank's http://www.amazon.com/Taming-Data-Tidal-Wave-Opportunities/dp/1118208781/ref=sr_1_3?s=books&ie=UTF8&qid=1354001370&sr=1-3&keywords=pmml. I think that Knime can convert R models into PMML (again I don't know a citation). Augustus also reads and writes PMML (http://rgrossman.com/2011/12/30/pmml-version-4-1-and-augustus-version-0-5-released/), but a better citation is needed. The 2nd edition of the book "PMML in Action" might be an appropriate source for some of the points made in this section: http://www.amazon.com/PMML-Action-2nd-Edition-Unleashing/dp/1470003244. Karl (talk) 07:35, 27 November 2012 (UTC)