KDnuggets : Newsletter : 1999 Issues : 99:10 Contents :

KDnuggets 99:10, item 11, Tools/Services:

Previous | Contents |  Next

Date: Mon, 26 Apr 1999 09:34:26 +0200
From: knowledgeminer@iworld.to (Frank Lemke)
Subject: Self-organizing Data Mining Using KnowledgeMiner

Self-organizing Data Mining Using KnowledgeMiner

KnowledgeMiner provides most objective, easiest, and fastest data mining 
available today. It works using three advanced self-organizing modeling 
technologies: Group Method of Data Handling (GMDH Neural Networks), Analog 
Complexing and Fuzzy Rule Induction. This is the first time that all of these 
algorithms have been available in one place on any computer platform.

Built on the cybernetic principles of self-organization, KnowledgeMiner can 
learn completely unknown relationships between the outputs and inputs of a 
given system in an evolutionary way from a very simple organization to an 
optimally complex one by itself very fast. For Neural Networks this means,
e.g., that it creates an optimal network topology automatically without 
predefinition. Also as it learns about a new dataset it generates equations, 
patterns or rules on the fly that model that data.

KnowledgeMiner has been used successfully for analysis, prediction and 
classification in the fields of finance, economics, imaging, ecology,
health, biotechnology, chemistry, math and many others. Professionals in
many areas can benefit by downloading and trying out the examples then using 
this tool to mine their own field for new relationships.

KnowledgeMiner can be downloaded from:
http://www.scriptsoftware.com/km/index.html
Included datasets and examples range from prediction of global
temperature, 
stock market trends, medical diagnosis, failure of materials (like the 
Challenger Space Shuttle O-Ring), to party affiliation in the US congress. 
The many examples included with KM show its power to work on human issues.

For more info and a free download see:
http://www.scriptsoftware.com/km/index.html

Previous | Contents |  Next


KDnuggets : Newsletter : 1999 Issues : 99:10 Contents :

Copyright © 1999 KDnuggets