Clementine’s open architecture makes it possible to add a complete in-database mining platform and visual workflow interface to your IBM DB2 Data Warehouse Edition database. Integrated capabilities for the entire data mining process capitalize on DB2 performance capabilities to increase data mining productivity and returns. Unlike other data mining solutions—which lack true in-database mining support—Clementine performs critical data preparation, modeling, and model scoring tasks within DB2.
Click image to enlarge.
Clementine scales data mining in DB2 using a three-tiered architecture:
Preparing data often accounts for 80-90 percent of the typical data mining process. Clementine helps minimize data preparation time and network traffic by enabling you to directly access and prepare data within DB2 databases, rather than on server or client tiers. Other data mining solutions force you to prepare data within their proprietary formats�adding unnecessary data preparation steps.
Clementine’s intuitive, visual workflow interface makes rapid "train-of-thought” modeling possible, so you stay focused on solving business problems rather than performing technical tasks, such as writing code. DB2 Intelligent Miner Modeling techniques are available as nodes within Clementine’s visual interface alongside native Clementine modeling techniques. All of these modeling techniques work together so you can quickly build and test numerous models to see which one produces the best result. Clementine provides the best and broadest analytics available—and an open design to extend further—so you have all the techniques necessary to deliver business value within a single, productive workbench.
Take advantage of integration with DB2 Intelligent Miner Visualization to easily evaluate Clementine association models within Intelligent Miner Visualization�s results browser. Clementine shares predictive models with Intelligent Miner Visualization using the emerging industry-standard XML mark-up language for data mining, Predictive Model Markup Language (PMML).
Strategic model deployment is the key to data mining ROI. Clementine provides two rapid deployment methods for scoring within DB2 databases—PMML and bulk loading—both of which have been validated by IBM. PMML models are easily deployed into DB2 databases for in-database scoring, without requiring Clementine Server processing, using DB2 Intelligent Miner Scoring.
Clementine's direct interface to DB2's bulk loading capabilities also provides the ability to quickly load data mining results, such as scores or segments, into your DB2 database. These in-database scoring features utilize the DB2 database's performance capabilities to accelerate the scoring of new cases and deliver predictive insight within daily business operations.
White Paper: Clementine Data
Mining Workbench
Bloor Research
praises Clementine's
leadership in data
mining market.
Key Products and Services
Resources
Buy
Demos and Downloads
