How to set up a non-english Stop List for Text Analysis

AndiWindisch — Tue, 04 Feb 2014 19:09:07 GMT

Hey,

we are very new to SAS Visual Analytics and have just upgraded our environment to 6.3 .

I've tried to load another 'Stop List' for the Text Analysis Feature, which is came with 6.3.

But i have no sucess.

First i created the Dataset for the 'Stop List', with german stop words named GERSTOP and only one variable 'TERM'. I load it to the LASR Server, into the same folder and library where the data for the text analysis is.

That's how i understood the manuel (SAS(R) Visual Analytics 6.3: User's Guide), but i won't work. I always get a information on the Word Cloud that no 'Stop List' ist defined.

Does anyone knows what is wrong in this setup? Other features of 6.3 will work.

Thanks a lot for you help!

Andreas

Re: How to set up a non-english Stop List for Text Analysis

Nascif_SAS — Tue, 04 Feb 2014 19:29:16 GMT

Hi Andreas,

You need also to load your new stop list in the Visual Analytics Data Builder by clicking Tools\Load Text Analytics Stop List: then accept the defaults when they come up.

Note however that currently VA Text Analytics supports only the English language. Changing the stop list won't help with that because there are language-specific features of the text analysis (identifying entities, dealing with declination, etc) that are not affected by that change.

Regards,

Nascif

topic How to set up a non-english Stop List for Text Analysis in SAS Visual Analytics

How to set up a non-english Stop List for Text Analysis

Re: How to set up a non-english Stop List for Text Analysis