Data Storage Formats

Poll

What are your preferred methods for storing data for data mining? [403 votes total] Text, CSV (comma-separated) (72) 18% Text, space or tab separated (55) 14% Excel (38) 9% SAS (57) 14% SPSS (31) 8% S-Plus/R (15) 4% Weka ARFF (23) 6% Other data mining tool format (11) 3% In a database system (93) 23% Other - please comment (8) 2%

Comments

Katharina Morik, Data together with KDD process meta data
The data themselves are, of course, stored in a database. However, if a data mining tool (e.g., MiningMart) which directly accesses the database also allows to store meta-data on the KDD process, as well, it combines the advantages of database formats (easy upload from transaction database, scalability, administration, and security) with tool formats (learning operators available, documentation of parameter settings,...).

Will Dwinnell, Data Mining File Formats
I also use MATLAB-format (.mat) files.