Awhile back somebody asked if there was any data-mining software for
CL. I was trying to get this out the door before his message
disappeared from my news server, but I guess I didn't make it. The URL
below contains a CL implementation of Peter Clark's CN2 rule induction
algorithm. CN2 is a modified version of the AQ algorithm already
available in the CMU AI repository. CN2, unlike AQ, can cope with noisy
data. What I've written is a bare-bones implementation of the simpler,
ordered version of the algorithm (it isn't fast, either). There are
pointers to other versions and papers at the site. Anyway, here's the
URL:
http://www.infinet.com/~btobin/dm.html
Bruce Tobin <······@infinet.com> wrote:
>Awhile back somebody asked if there was any data-mining software for
>CL. I was trying to get this out the door before his message
>disappeared from my news server, but I guess I didn't make it. The URL
>below contains a CL implementation of Peter Clark's CN2 rule induction
>algorithm. CN2 is a modified version of the AQ algorithm already
>available in the CMU AI repository. CN2, unlike AQ, can cope with noisy
>data. What I've written is a bare-bones implementation of the simpler,
>ordered version of the algorithm (it isn't fast, either). There are
>pointers to other versions and papers at the site. Anyway, here's the
>URL:
There's an extension of XLisp called XLispStat, developed by Luke
Tearney. It's free to download but the book costs �65 (sixty five
English pounds).
See http://stat.umn.edu/~luke/xls/xlsinfo/xlsinfo.html for more info.
You can certainly use XLispStat to do exploratory data mining. It
lacks a lot of the statistical algorithms built into S-Plus or SAS.
Unlike those, it's a general-purpose programming language with some
statistical support, rather than a statistical analysis package with
some programming support.
> http://www.infinet.com/~btobin/dm.html
Le Hibou http://homepages.enterprise.net/hibou/
"What the ... This is Lambic! Where's my culture of amoebic
dysentery?" -- Gary Larson