KDnuggets Data Mining Poll 2007

The KDnuggets poll has a dramatic showing for Clementine fans. 116 of 534 respondents use Clem.

 http://www.kdnuggets.com/polls/2007/data_mining_software_tools.htm

Perhaps more striking is that it the strongest performer of those that use it as their only tool, so it earns high scores for comprehensiveness. A finding that I have seen in this poll before and that has puzzled me is Excel's appearance at number three after Clem and Salford systems. There is really no mystery, however. Only two respondents claim to use Excel exclusively. I literally can not imagine what those two folks can accomplish in only excel. Almost certainly the others are mining in one package, but reporting part of their results in excel. This excel finding is a near duplicate of the year before: http://www.kdnuggets.com/polls/2006/data_mining_analytic_tools.htm

Like most of my colleagues that use SPSS, I am agnostic on SAS. I wish I were "bilingual", so to speak. Remarkably, only 8 out of 534 respondents use either SAS product exclusively for data mining. I don't know if this reflect the real world, or merely reflects the KDnuggets crowd.

This year they broke out the free software separately with WEKA, Yale, and R performing the best. See my recent post regarding R.

For Yet Another Learning Environment (YALE): http://rapid-i.com/

For WEKA: http://www.cs.waikato.ac.nz/ml/weka/