Zementis

Zementis is a leading software company focused on the operational deployment and integration of predictive analytics and data mining solutions.  Its products are designed from the ground up to benefit from open standards and to significantly shorten time-to-market for predictive analytics in any industry.

 

Zementis

Benefits of Partnership

The partnership between Teradata and Zementis allows customers with a variety of data mining tools to efficiently deploy predictive models based on the Predictive Model Markup Language (PMML) standard.  Focused on Big Data applications, the Universal PMML Plug-in for Teradata enables scalable execution of standards-based predictive analytics directly within your Teradata data warehouse.

Benefits of the Teradata-Zementis Partnership

  • Performance enhancement: The UPPI integration deploys a general purpose scoring accelerator where the customer can use any PMML compliant model building environment (e.g. R) and target a database platform as the production scoring environment. This allows the customer to leverage their investment and expertise in their current set of data mining tools and get productivity and performance increases right-away. In addition, UPPI allows for companies to standardize their predictive projects on a common language. There is no need to learn yet another tool or environment.
  • Use best-of-breed tools: Since PMML is an open-standard, it allows for all a customer’s predictive solutions to be represented in a single and concise way. Given that all the top model building tools support PMML, it allows Teradata to enable customers to use a best-of-breed approach to model development. In this case, even if one uses SQL-MapReduce® to build a tree model, but R to build a Random Forest model, both models are represented in PMML and deployed right away using UPPI.
  • Fosters transparency and best-practices: PMML also allows for transparency and for spreading best-practices among analytics teams. With PMML, predictive models are not boxed in into a single solution or represented in a cryptic format. Instead, they are represented in a common standard that can be easily used and understood.
  • No custom coding necessary: With the PMML integration, there is no need for custom coding. Models are represented in a single language and moved around between systems as such. In this way, after a model is built, its PMML representation just needs to be moved into UPPI to be deployed. With PMML and UPPI, model deployment takes minutes, not months. This derives the ultimate benefit of immediate insights and increases the likelihood of deploying solutions to fill gaps in key areas of business development.

 

Universal PMML Plug-in (UPPI) for Teradata

UPPI for Teradata delivers in-database scoring for rapid, on-the-fly predictive analytics on large volumes of data. UPPI allows businesses to deploy predictive analytic models based on the PMML standard. Analysts can quickly integrate a wide range of analytic models generated by leading commercial and open-source data mining tools into the database for optimal scalability and performance.

Quick Facts

VersionTeradata Version(s)
3.6TD 14.0
TD 13.10

Product Resources

UPPI for Teradata Product Data Sheet
Universal PMML Plug-in (UPPI) for in-database scoring of predictive analytics. Product Data Sheet.

Universal PMML Plug-in (UPPI) for Aster

UPPI for Aster delivers in-database scoring for rapid, on-the-fly predictive analytics on large volumes of data. UPPI allows businesses to deploy predictive analytic models based on the PMML standard. Analysts can quickly integrate a wide range of analytic models generated by leading commercial and open-source data mining tools into the Aster database for optimal scalability and performance.

Quick Facts

VersionTeradata Version(s)
3.6AD 5.0.1

Product Resources

UPPI for Teradata Aster Product Data Sheet
Universal PMML Plug-in (UPPI) for in-database scoring of predictive analytics. Product Data Sheet.