Data mining tools you regularly use

Poll

Data mining tools you regularly use: [967 choices, 551 voters] SPSS Clementine (128) 13% Weka (101) 10% SAS (100) 10% CART/MARS (89) 9% SPSS/AnswerTree (76) 8% SAS Enterprise Miner (67) 7% Other commercial tools (65) 7% Other free/open-source tools (57) 6% MATLAB (52) 5% Microsoft SQLServer/Excel (40) 4% Insightful Miner (36) 4% IBM Intelligent Miner (35) 4% KXEN (35) 4% C4.5 / C4.8 (29) 3% Angoss (26) 3% Megaputer Polyanalyst (10) 1% Neuralware (8) 1% Oracle Suite (Darwin) (8) 1% Quadstone (3) 0.3% ThinkAnalytics (2) 0.2%

Comments

• Javier Pison, Jun 16, 2002; Subject: 'R' and 'WEKA'
  I use open source "R" (www.r-project.org) and "WEKA". "R" is a powerfull statistical tool to manipulate the data and "WEKA" is very useful to analize them.
  Congratulations for all the people who have developed "R" and "WEKA".
  
  
• Larry M Deschaine, Jun 9 2002; Subject: Scientific Data Mining
  I have had great success writing equations from numeric data with a multiple run, linear genetic programming system called Discipulus [not heavily marketed and extremly effective]. It is distributed by RMLT, www.aimlearning.com. Some of the published work I have used it for is found on my Home Page: http://pw2.netcom.com/~lmdmit84.
  
  The tools from Salford systems are also quite impressive and deserve a close and serious look if you have not used them yet.
  
  I get many emails, and will try to respond if requested, and I do ask that you review the information on the web sites first - thank you in advance for your collaboration - Larry
  
  
• Will Dwinnell, Jun 7, 2002; Subject: MATLAB
  I have found MATLAB to be a tremendously useful tool. It is a natural fit for numerical work. Implementing algorithms is a breeze compared to languages like C/C++, etc. Pre- and post-processing are also a good fit.
  
  
• Sheldon Evans, Jun 6, 2002; Subject: Statistica Data Miner
  I'm an early adopter of a product called Statistica Data Miner. It's far and away the most comprehensive data mining solution on the market. It gives me the freedom to customize while including features that other products such as Clementine and Enterprise Miner don't. Their enterprise version is much more cost effective than SAS, IBM, and SPSS.
  
  
• Ted Pedersen, Jun 5, 2002; Subject: why I like weka
  I use Weka quite a bit and recommend it highly. It is open source, easy to use, and quite extensive in it's coverage. It's also free, and the developers have proven very helpful and provide quite good support.
  
  
• prasanna, Jun 5, 2002; Subject: Weka
  The best thing I like about Weka , is that it is open source , user is free to do modifications in the source code , easy to use GUI features .. Weka must be already giving hiccups to high priced data mining products..
  
  
• Aldebaro Klautau, Jun 4, 2002; Subject: Weka
  I like Weka because it is very easy to use is one wants to simply run simulations with standard algorithms. And also, when I feel like implementing new and complicated non-working algorithms myself, Weka's documentation is very nice (e.g. references to papers) and the code is in Java (a nice language that only Gates dislikes).
  
  
• Liu Huiqing, Jun 4, 2002; Subject: I like Weka
  Weka is a very nice open-source data mining tool. It helps me greatly.
  
  
• Nicholas, Jun 4, 2002; Subject: You Forgot BioComp
  You forgot BioComp's iManageData, iModel, iUnderstand and iImprove.
  
  
• Hercules Prado, Jun 4, 2002; Subject: My tool of choice for classification.
  For classification tasks I use AIRA, a tool based on Combinatorial Neural Model, developed by Hycones.