“As a standard, PMML provides the glue to unify data science and operational IT. With one common process and standard, PMML is the missing piece for Big Data initiatives to enable rapid deployment of data mining models. Broad vendor support and rapid customer adoption demonstrates that PMML delivers on its promise to reduce cost, complexity and risk of predictive analytics,” says Alex Guazzelli, Vice President of Analytics, Zementis. “You can not build and deploy predictive models over big data without using multiple models and no one should build multiple models without PMML,” says Bob Grossman, Founder and Partner at Open Data Group.
Some of the elements that are new to PMML v4.2 include:
- Improved support for post-processing, model types, and model elements
- A completely new element for text miningScorecards now introduce the ability to compute points based on expressions
- New built-in functions, including “matches” and “replace” for the use of regular expressions
“With PMML our customers and partners are able to drive real value from their predictive models right away, using the open standard,” said Andrew Flint, Senior Director of Product Management at FICO (NYSE: FICO). “Models built in most commercial or open source data mining tools, such as FICO Model Builder or R, can be instantly deployed in the FICO Analytic Cloud using PMML. The net result is quicker time to innovation and value on analytic applications, and the ability to combine the power of standards-based predictive analytics with the scalability of cloud computing.”
Radhika Kulkarni, Vice President of Advanced Analytics at SAS notes, “SAS continues to support the analytic collaboration that PMML provides to users. The recent release of SAS Enterprise Miner 13.1 provides users the ability to not only consume PMML from Open Source R models, but also produce PMML, which can be consumed by other applications. SAS Model Manager enables users to consume and manage R and PMML models as part of the SAS ecosystem. Sharing analytic models is paramount to the analytic lifecycle.”
“We are extremely excited to continue our long-running PMML support to PMML 4.2,” said Scott Cappiello, Vice President, Program Management, MicroStrategy Incorporated. "As a firm believer in providing the maximum analytic flexibility to organizations, PMML provides significant advantage in folding in analytics beyond our native and open source R capabilities, to provide business users the full range of business analytics in the big data age”.
"We are happy to see PMML's impact continuing to grow and will keep being among the first to integrate new PMML features into KNIME. Starting with our summer release KNIME will also support Naive Bayes Models and we will keep adding to its PMML Preprocessing abilities as well," says Kilian Thiel of KNIME.
PMML is the leading standard for statistical and data mining models and supported by over 20 vendors and organizations. With PMML, it is straightforward to develop a model on one system using one application and deploy the model on another system using another application.
The Data Mining Group (DMG) is an independent, vendor led consortium that develops data mining standards, such as the Predictive Model Markup Language (PMML). DMG members include: IBM, MicroStrategy, SAS, Experian, Pervasive Software, Zementis, Equifax, FICO, KNIME, NASA, Open Data Group, Rapid-I, Togaware, and Visa.
For more information about the Data Mining Group and the PMML standard, go to: www.dmg.org