Being at KDD (Knowledge Discovery and Data Mining) right now - not to mention having just sat down for a chat with Karl Rexer, I thought it fitting to post a summary that Karl shared of his recent data mining survey:

2007 HIGHLIGHTS:

·   27-item survey of data miners, conducted on-line in early 2007

·   314 responses from individuals in 35 countries

·   Regression, decision trees and cluster analysis were the most commonly used algorithms (mean number of algorithms used: 6.8)

·   Top challenges data miners report are dirty data, data access, and explaining data mining to others

·   SPSS, SPSS Clementine, and SAS are the three most frequently utilized tools (mean number of tools used:

4.5)

·   There is increasing interest in the Oracle Data Mining tool, and decreasing interest in C4.5/C5.0/See5   

·   The primary factors data miners consider when selecting an analytic tool are: 1) the dependability and stability of software, 2) the ability to handle large data sets, and 3) data manipulation capabilities

·   The findings vary somewhat depending on the domain in which the data miner works, the tools used, geography, and several other dimensions