Attributes names and import process

greg

About attribute names, they seems to be very "old school" (only characters, like a regular programming language), am I right?
The "import" command however is not as rigid, so I end up with attributes containing spaces and accentuated chars.

So my question is: when I have a repository created by he "import" command, and has bad attribute names, is it possible to "correct" the repository without re-doing the import process?

TIA

greg

PS: BTW I don't know if it's a known bug, but on my system importing an excel sheet only works the first time after launch (meaning to import another one I have to quit and relaunch rapidminer)

[Deleted User]

Hi greg,

First of all let me state that RapidMiner is capable of dealing with all character sets since you can also specify the file encoding within the import wizard.

You can use the Operator Rename which gives you the possibility to rename all attributes within one operator.

Please keep in mind that you should not use blanks or hyphens within the new names if you want to compute e.g. the difference between the values of those attributes.

Best,

Edin

greg

Thanks for your answer.

This is not exactly what I'm looking for, but it could help. My point was about changing the import parameter of a datasource created by the "import" menu item. Maybe I need to pratice more then I'll see if I really need it.