Continuing the trend of packing BI (business intelligence) functionality into a database engine, IBM Corp. has added a piece of its Intelligent Miner Scoring data mining tool to its relational database DB2 Universal Database.
The services of the currently available Intelligent Miner Scoring will now enable real-time data gathering capabilities within the core database, said Janet Perna, general manager of data management solutions at IBM.
"This is where we see database technology going," Perna said. "We are going to see more and more of this support inside the database."
Indeed, both Oracle and Microsoft have pulled data mining and BI capabilities into their core relational database offerings.
"The database vendors are finally starting to get it and are adding functionality to help end-users use the data," said Howard Dressner, an analyst at Gartner Group. "By putting the intelligence close to the data, users, when they interact with these engines, can get more insight out of the data."
Although IBM is not doing anything "earth-shattering, it is certainly good for users," Dressner said. "Data mining is becoming part of the database infrastructure."
IBM's Perna added that the data mining services enable companies to dredge data for predictive modeling, such as customer buying behavior.
For instance, if a customer interacts with a call center, the software can automatically score the caller to understand the probability that he or she will buy a specific product.
The data mining functionality can also be integrated with third-party applications. Big Blue will ship a cartridge to extend data mining services to Oracle databases, Perna said.
Mellon Bank has been using Intelligent Miner for three years, as well as several other mining tools and databases from IBM and Oracle, said Jim Delaney, vice president of strategic technology at Mellon. "Bringing data mining into the database eliminates the middleware functionality," Delaney said. "It is no longer required if the database and the tools are right there."
Delaney continued to say that, in addition to not having to pay for and manage the previously necessary middleware, another advantage seen by Mellon Bank was predictive modeling based on the industry standard, PMML (Predictive Modeling Markup Language).
The data mining services will ship in DB2 Universal Database at the end of this month.
Mining for gold in relational databases
IBM, Microsoft, and Oracle have all recently grafted in data mining and BI capabilities to provide benefits.
* BI accessible to more users
* Data easier to analyze
* Analytics near real time with better performance