|
GUHA (General
Unary Hypothesis Automaton) is one of data mining methods
(basicaly a method for exploratory data analysis), which enables the recognition of
regularities in empirical data.
Its aim is to offer "all interesting" hypotheses supported by given data and differs
from many statistacally based data mining methods by the emphasis on polyfactorial
hypotheses searching.
In the last 30 years different GUHA versions have been successfully applied in
the following areas: medicine, sociology, psychology and reliability studies in the
automotive industry.
Now there is new implementation of the GUHA method
for Windows 95/98/NT platform available.
This implementation called as GUHA +- project can be downloaded
here
(to unzip use -d switch). Software is licensed as freeware. Instalation
includes reference manual. Another more user friedly manual including description
of GUHA +- package installation is
here as zipped postscript or
here as zipped MS Word document.
For other info mail to coufal@cs.cas.cz
HTML version of the
PC-GUHA Brief manual
is also available. PC-GUHA is an old implementation of GUHA method for DOS.
It will be also available for download, but now it is under process of recompilation
so check this site in future.
|