[R] PMML- Predictive model markup language

A.J. Rossini rossini at blindglobe.net
Sat Feb 24 19:38:12 CET 2001

>>>>> "JA" == John Aitchison <jaitchis at lisp.com.au> writes:

    JA> I caught a passing reference to PMML on the WEKA list (I
    JA> think) .. an XML way of communicating 'data mining models'.  I
    JA> am of the opinion that XML is a 'good thing' and I would for
    JA> instance like to see a standard XML way of communicating the
    JA> content and context of datasets (eg of timeseries) but it has
    JA> been a long day and my internet connection is slow and so I
    JA> have not worked through the slideshow at http://www.dmg.org/
    JA> but wonder if anyone else has examined PMML and can give an
    JA> opinion as to whether this (PMML) is worthwhile opening up as
    JA> a discussion thread.

I looked at it in the past, and it is quite intriguing -- it's similar
to some work people are doing on literate data/statistical analysis
(round-trip).  However, it's more specific and focused, and I'm not
sure that I agree with how they are communicating models (but that's a
small disagreement on specific implementations, not on the need for
this type of communication system for data analysis).


A.J. Rossini				Rsrch. Asst. Prof. of Biostatistics
UW Biostat/Center for AIDS Research	rossini at u.washington.edu	
FHCRC/SCHARP/HIV Vaccine Trials Net	rossini at scharp.org
-------- (friday is unknown) --------
FHCRC: M--W : 206-667-7025 (fax=4812)|Voicemail is pretty sketchy
CFAR:  ??   : 206-731-3647 (fax=3694)|Email is far better than phone
UW:    Th   : 206-543-1044 (fax=3286)|Change last 4 digits of phone to FAX
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list