Chapter 1. Predictive Model Markup Language (PMML)
Predictive Model Markup Language (PMML) is an XML-based standard established by the Data Mining Group (DMG) for defining statistical and data-mining models. PMML models can be shared between PMML-compliant platforms and across organizations so that business analysts and developers are unified in designing, analyzing, and implementing PMML-based assets and services.
For more information about the background and applications of PMML, see the DMG PMML specification.
1.1. PMML conformance levels
The PMML specification defines producer and consumer conformance levels in a software implementation to ensure that PMML models are created and integrated reliably. For the formal definitions of each conformance level, see the DMG PMML conformance page.
The following list summarizes the PMML conformance levels:
- Producer conformance
- A tool or application is producer conforming if it generates valid PMML documents for at least one type of model. Satisfying PMML producer conformance requirements ensures that a model definition document is syntactically correct and defines a model instance that is consistent with semantic criteria that are defined in model specifications.
- Consumer conformance
- An application is consumer conforming if it accepts valid PMML documents for at least one type of model. Satisfying consumer conformance requirements ensures that a PMML model created according to producer conformance can be integrated and used as defined. For example, if an application is consumer conforming for Regression model types, then valid PMML documents defining models of this type produced by different conforming producers would be interchangeable in the application.
Red Hat Decision Manager includes consumer conformance support for the following PMML 4.2.1 model types:
For a list of all PMML model types, including those not supported in Red Hat Decision Manager, see the DMG PMML specification.