DISCO <data>,<output line> / Reijo Sund 2001
Knowledge DISCOvery tools for Survo
Performs some data mining operations on active string
variables in <data>. The order of variable blocks must
be specified by BLOCKORDER=<activation letters> and the
operation method by METHOD=<EXTRACT,REMAP,FIND>. The upper
limit of different boolean items in variables can be given
by MAXITEMS=<number of items>.
EXTRACT-method scans throught data and counts the
occurences of different items in variables. The items are
written to a text file if the output file is specified by
ITEMOUTFILE=<file>.
Using REMAP-method it is possible to "rename" the items in
data. However, the original data is not altered - the name
for output (text) file must by given by REMAPOUTFILE=<file>.
The remapping rules - of the same form as EXTRACT-method's
output - are specified by REMAPFILE=<file>.
FIND-method performs level-wise frequent itemset counting
for boolean items (See e.g. Han & Kamber (2001): Data Mining:
Concepts and Techniques, chap. 6). The minimum frequency
threshold can be specified by MINFREQ=<threshold frequency>.
The results are written to a text file if specification
FREQOUTFILE=<file>. The ITEMOUTFILE-specification can be
used as in EXTRACT-method.
M = More information on multivariate analysis
Help System (web edition)