SPSS delivers Data Mining

Share this content

In an effort to deliver on the promise of scalable data mining products, SPSS is now shipping server versions of its flagship products - SPSS Server 10.0 and Clementine Server 5.1. Both software products are designed to enable enterprises to analyze large centralized data stores and deploy data mining models to decision makers throughout their organization. These releases are considered by SPSS to be "milestones" in an aggressive plan to make the company's business intelligence products better suited to handle the analysis needs of large-scale organizations.

"Our development plan is aimed at addressing businesses' growing need to efficiently and effectively leverage the tremendous amounts of data they accumulate daily," remarked SPSS President and CEO Jack Noonan. "SPSS Server and Clementine Server enable people to perform valuable analysis of huge, centralized data sources more quickly, while reducing network traffic and maintaining data security." He added, "These advances in analysis fulfill the first step in getting greater strategic benefit from data. The next step requires deploying the models built during analysis to decision makers throughout the enterprise."

SPSS Server and Clementine Server have been designed to make use of system resources for the types of analyses they perform. For example, SPSS Server eliminates the need to make copies of the source data or convert it to SPSS format through a Virtual Active File. The Virtual Active File keeps track of any changes to the data, such as data transformations, and records only the difference between the new data and the original data, removing the need to copy unaffected variables or columns. Because copying and conversion of the data are eliminated, SPSS Server is able to analyze large amounts of data in short amounts of time. Users can also adjust the Virtual Active File to achieve the ideal balance between the amount of disk space used and analysis performance.

Clementine Server pushes back into the database as many functions (such as sorts, joins, aggregations, etc.) as possible to speed performance and reduce network traffic. These changes reduce the amount of time needed for data manipulations in each step of the data mining process.

To Make all of its business intelligence products scalable is part of SPSS' plan to help organizations gain insight into their data. The second part of that plan is to provide creative mechanisms to deploy results throughout organizations.

Other business intelligence products scheduled for 1999 release include systems for analyzing time-based data and deploying the resultant models and  for report distribution.

About mycustomer.newsdesk


Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.