Features

The previous KDnuggets poll asked

What dataset format you use the most when data mining?

Based on 238 votes, about half of the responders chose either comma-separated of tab-separated formats. Another quarter of the responders indicated that they kept the data in a database. Only 11% kept the data in a commercial data mining tool format such as SAS.

Full results:

Comma-separated (.csv) file		24%
Tab or Space-separated (.txt) file	26%
Commercial data mining tool format	11%
Weka format (.ARFF)			 5%
in a database				27%
in a spreadsheet			 7%
Other					 1%