Fog Creek Software
Discussion Board




open source data mining -info retreival packages

Anybody aware of any open source packages for classification, clustering and info retreival, which could be used in the enterprise?

Sumit
Friday, July 16, 2004

Free as in I'm a cheap bastard?

.net, the equivalent of MS Bob.
Friday, July 16, 2004

www.dwreview.com - used to have an open source data mining, and OLAP, tool for MySQL

We have used it in our college for the last two years.

It does not have fancu GUIs like SAS but it works well and has tons of APIs.

But I cannot find the download on the site any more - the author is Hari Mailvaganam - maybe he has it on source forge or else where.  The same team also worked on the data mining module at Microsoft.

Hope this is helpful.

Wierd Al Yankovic
Friday, July 16, 2004

http://sourceforge.net/projects/dat-a/

mike
Friday, July 16, 2004

http://www.prudsys.com/Produkte/Algorithmen/Xelopes

Peter
Friday, July 16, 2004

We have experimented with mining using the Orange package. http://magix.fri.uni-lj.si/orange/ Seems to be working well for our needs so far.

g
Friday, July 16, 2004

We have been using R for all our data mining and statistical computing needs, and have been very satisfied with the results.  http://www.r-project.org/ (or try the US mirror http://cran.us.r-project.org/)

We are focusing more on Customer Intelligence analytics, and have published some of our models into the open source community as well. See R.loyaltymatrix.com for more information on our implementation of R

hth

Sandeep Giri
Wednesday, August 04, 2004

*  Recent Topics

*  Fog Creek Home