K.wiz by thinkAnalytics Corporation

0.90 - What's this?

K.wiz is a new style of Knowledge Discovery product, supporting the entire KD process to scale and deploy to enterprise requirements for such applications as customer segmentation and churn.

English

Supported Technologies

Windows 95/98/ME, Windows XP/2000/NT , HP/UX, Solaris/Sun OS
Web-Based (Browser)
Click on a technology to view similar products within this category.

Pricing

Users (# of seats)
30,000 to 100,000
info@thinkanalytics.com
617 331 8364

End user, VAR and OEM agreements welcome, reseller and consulting packages

Additional software product description, benefits, features, and uses.

Additional Product Information

K.wiz delivers a new Knowledge Discovery (KD) solution. Combining;ease of use with Scalability, this advanced client server framework;encompasses all stages of the KD process. K.wiz components deliver;data transformation, visualisation, and discovery algorithms to;the desk and Web top. External Components extend the already powerful;range of functionality and ensure K.wiz fulfils each organisation's;unique demands. K.wiz is written entirely in Java and based on the;latest 1.2 version. The K.wiz framework provides a scaleable;distributed architecture that enables machine learning and data;mining algorithms to exploit the power of the Java platform, bringing;an unbeatable edge to the realm of business intelligence in a;networked world.;Major problems in supporting the KD process involve the appropriate;and timely supply of massive data sets, the storage of and access to;interim results during the process, and the storage and access to the;result data sets. It is a common criticism that database vendors;make no allowance for, or understanding of, the requirements for;data mining access. Components available in K.wiz address these;issues by tuning and providing novel access functionality most;suited for data handling in a KD process. The database, buffer and;cache management systems used throughout K.wiz use entropy based;compression techniques to take full advantage of low cardinality;data which have been benchmarked to show that compression of the;order of 4% of original size can be realised.


Search within this category