|
|
 | | From: | Tim Menzies | | Subject: | data structures for rule-covering algorithms | | Date: | Thu, 20 Jan 2005 17:19:06 GMT |
|
|
 | so i've been writing a PRISM/INDUCT clone that
finds attribute ranges of interest, removes
all instances with those ranges, then repeats
on the remaining data.
i had a "good idea" of keeping indexes attribute values
for each instance and incrementally updating those
indexes when i removed some data.
this has proven to be surprisingly complex so i was
wondering if anyone had a view on good data structures
for the internals of a rule covering algorithm?
tim menzies
http://menzies.us
[ comp.ai is moderated. To submit, just post and be patient, or if ]
[ that fails mail your article to , and ]
[ ask your news administrator to fix the problems with your system. ]
|
|
|
