PMML FAQ
Q. What is Predictive Model Markup Language (PMML)?
Predictive Model Markup Language (PMML) is an XML mark up language to describe statistical and data mining models.
Q. How would you use Predictive Model Markup Language (PMML) to describe a predictive model?
PMML describes the inputs to data mining models, the transformations used to prepare data for data mining, and the parameters which define the models themselves.
Q. How is PMML used?
A. PMML is used for a wide variety of applications, including applications in finance, e-business, direct marketing, manufacturing, and defense. PMML is often used so that systems which create statistical and data mining models ("PMML Producers") can easily interoperate with systems which deploy PMML models for scoring or other operational purposes ("PMML Consumers").
Beginning with PMML Version 2.1, PMML is also beginning to be used to describe the process of preparing data for data mining. PMML Version 3.0 will provide additional functionality for describing the transformations, aggregrations, and normalizations which are needed to prepare data for data mining and statistical modeling.
Q. How can I get a current specification of the protocol?
The current release of PMML is Version 2.1. This is available via the source forge project pmml at www.sourceforge.net/projects/pmml.
Q. Who has released PMML products?
A. PMML is used in released products by SAS, SPSS, IBM, Microsoft, Oracle, and many other vendors. See www.dmg.org for a list of vendors supporting PMML. It is the most widely deployed data mining standard.
Q. How do I join the Data Mining Group?
A. There are two types of memberships. Vendors shipping products containing three or more statistical and data mining models are elgible for full membership. Other interested parties are elgible for associate membership. Membership requires approval by the Data Mining Group.
Q. Is the Data Mining Group part of any standards group?
A. PMML is part of xml.org.
Q. What is the development history of PMML?
PMML 0.7 was released by the
National Center for Data Mining (NCDM) in July 1997.
PMML 0.9 was developed by the Data Mining Group and released in July, 1998.
PMML 1.0 was released in August, 1999.
PMML 1.2 was released in August, 2000.
PMML 2.0 was released in August, 2001.
PMML 2.1 was released in March, 2003.
PMML 3.0 is planned for release in August, 2004.