Data warehouse giant Teradata on Dec. 13 introduced a new suite of data-mining software for enterprises that it says significantly automates data preparation and accelerates performance of its partners data-mining tools.
Teradata Warehouse Miner 5.0 lets users advance beyond simply describing what their customers did last quarter to predicting future buying behavior, a company spokesperson said.
“This innovative technology enables companies to convert information into insight, and insight into actions which support better, faster decisions,” said Randy Lea, vice president for Dayton, Ohio-based Teradatas products and services.
Teradata Warehouse Miner 5.0 has been designed to optimize analytics when deployed in combination with SAS, SPSS, Fair Isaac Model Builder or KXEN data-mining tools, Lea said.
“The combination of the Teradata in-database performance with other vendors data-mining tools can speed model development as much as two to three times faster and runtime of data-mining models up to 25 times faster,” Lea said.
“This in-database performance is extremely valuable to our customers and partners, because they use it to gain more insight, through more analytic models, executed against more customers data, across the organization in real time.”
Teradata Warehouse Miner 5.0 optimizes the data-mining process and can accelerate the use of partners data-mining tools in the following categories:
Data Profiling: Using the data exploration features, businesses can identify and resolve a wide range of data-quality issues that include: identifying duplicate records and missing data, validating data accuracy, verifying the format of data and identifying outlying data that may skew an analysis.
ADS generation: The Teradata ADS (Analytic Data Set) Generator is a flexible, open data-mining product that streamlines the most time-consuming and critical steps of data mining: the preparation of data for analysis. Most businesses spend up to 70 percent of their data-mining time and resources just getting data ready for analysis, which is a waste of valuable resources.
Extended Predictive Model Markup Language: Teradata has extended support of the XML-based PMML (Predictive Model Markup Language) to additional data-mining vendors. PMML is an open standard that allows businesses to share analytic models among applications that enable them to easily port desktop analytic models to leverage the power of a large parallel database with minimal effort. The use of PMML promotes performance and enables scalability.
Model Management: With Teradata Model Manager, Teradata has simplified the complex data-mining execution process by making it faster and easier for technical as well as nontechnical business users to convert massive amounts of detail data into analytic intelligence. Business users can now execute the analytic model against fresh data by targeting new customer or product segments.
A major European bank improved its target marketing campaigns with Teradata Warehouse Miner by better identifying potential banking customers with a 300 percent improvement in response over the control group, a company spokesperson said. By utilizing customer data, the bank understood how to focus its offers and customer service on fewer customers to produce better sales results.
A U.S. wireless provider increased contract renewals by 50 percent after it proactively contacted selected customers based on analytic intelligence that provided clues to when they were likely to terminate their wireless contract. This Teradata customer now has the lowest attrition rate of any wireless provider, the spokesperson said.
“The combination of KXEN and Teradata technology, and, in particular, Analytic Data Set Generator and Model Manager, opens the door to unlimited smart use of data to generate real insight and increased profitability for our joint customers,” said Roger Haddad, chairman and chief executive officer of KXEN.
Teradata Warehouse Miner 5.0 is available now. For pricing information, go here.