The Data Mining Group
, a vendor-led consortium involved in "developing standards for statistical and data mining models," has reportedly announced
the general availability of Version 4.0 of the Predictive Model Markup Language.
This new version of PMML is a major update of PMML Version 3.2, which was released May 2007.
The idea behind the product is to make it "straightforward" to develop a model on one system using one application and deploy the model on another system using another application. Group officials said it's supported by more than 20 vendors and organizations.
"PMML turns the deployment and practical application of predictive models within any existing IT infrastructure into reality,” said Cris Payne, senior analytics scientist for XO Communications (News
), Inc., in a statement. “Without PMML, it would take months for models to be integrated and deployed via custom code or proprietary processes.”
This version offers support for multiple models, which includes support for both segmented models and ensembles of models. Group officials said it also has "improved support for preprocessing data," designed to help with simplifying the deployment of models.
Oh, and 4.0 also has what group officials term "survival models." In case you were worried about that.
"PMML simplifies the deployment of analytic models," said Robert Grossman, chair of the Data Mining Group, in a statement. "With PMML, deploying a new model is as easy as reading an XML file."
Follow ITEXPO (News - Alert) on Twitter: twitter.com/itexpoDavid Sims is a contributing editor for TMCnet. To read more of David’s articles, please visit his columnist page. He also blogs for TMCnet here.
In July, TMCnet had the news that MicroStrategy (News
) Incorporated announced
that it had enhanced its predictive analytics capabilities with support for PMML 4.0.
"Data mining is an organic part of the MicroStrategy architecture that enables users to discover hidden patterns and predictive information in the data through standard enterprise reports and dashboards," company officials said, adding that while BI provides deep insight into historical data, "data mining uses this information to help companies forecast future events."
Edited by Amy Tierney