THE ASIA-PACIFIC CENTRE FOR RESEARCH, INC.

home > articles >

Latest version of SPSS data mining workbench adds Microsoft® SQL Server in database mining option and IBM® DB2® Data Warehouse Enterprise Edition in database scoring
Source: www.spss.com
Copyright SPSS, Inc. 2004
06/14/04

This is a dated announcement. The material in this announcement could be superceded by more current announcements.

Clementine® 8.5 adds new high speed association detection algorithm

CHICAGO, 05/19/04 — SPSS Inc. (NASDAQ: SPSSE), the leading provider of predictive analytics, today announced the immediate availability of Clementine® 8.5, the latest version of its award-winning data mining workbench. This new version leverages Clementine's open, standards-based architecture to offer easy integration with existing operational systems. Clementine 8.5 offers Microsoft® SQL Server 2000 Analysis Services users the ability to perform in-database mining, and an in-database scoring option for analysts using IBM® DB2® Data Warehouse Enterprise Edition. With these new options, Clementine can support full in-database mining and also model deployment in predictive markup language (PMML) that can be embedded in other systems and applications, such as IBM® DB2® Data Warehouse Enterprise Edition.

With ease of integration being the focus of Clementine 8.5, this latest version offers a variety of new deployment options. Analysts using Microsoft® SQL Server 2000 Analysis Services can now build Microsoft® Decision Trees directly within Clementine's intuitive visual workflow interface. Clementine 8.5 enables analysts using the Data Warehouse Enterprise Edition of DB2 to view Clementine association models with IBM's DB2 Intelligent Miner™ Visualization, and automatically deploy models built in Clementine into DB2 Intelligent Miner™ Scoring. This release also enables companies to author and deploy streams from Clementine into PredictiveCallCenter™, a new SPSS predictive analytic application that integrates with call center CRM and call management systems to instantly determine which inbound callers are good candidates for up-sell, cross-sell or retention offers.

Clementine, which supports the full data mining process and provides a visual map of data mining steps, offers a range of machine learning and statistical techniques for clustering, classification, association and prediction. This latest version incorporates CARMA, a high-speed association detection algorithm that delivers faster processing and gleans more detailed information from data. Additionally, Clementine 8.5 features a new association viewer that interactively filters, sorts and subsets association models enabling analysts to select the most relevant associations.

"Our aim is to make the analyst's job easier by helping them to work more quickly, efficiently and effectively. One of the ways in which Clementine achieves this is with its ease of integration, which enables organizations to take advantage of their existing hardware, software and data warehousing investments," said Laura Squier, Clementine product manager. "Clementine's open architecture leverages an organization's database investment by performing mining tasks including data access, pre-processing, scoring and now modeling, on the database tier, rather than on the server or client tiers. Analysts are able to save time, maximize resources and minimize network traffic."