"Problem with Sentiment Analysis"

lmcdowell

I am trying to run this wizard on a dataset in a CSV file with two columns - one is the Sentiment column and one is the Text column with open-ended survey responses. There are missing values in the Sentiment column to the process will have something to work with. However, I keep getting an Analysis Process Error stating "This application process cannot be executed on your data. Please make sure the structure of your data matches the one from the demo data. Also, make sure to select an appropriate column in Step 3." I am using the starter version of RapidMiner Studio 6.0.008.

fras

The name of the Text column must be "Text". Did you check that ?

lmcdowell

Sorry, didn't see this reply before. Yes, the second column is named Text.

Marco_Boeck

Hi,

I think most problems with the wizard come from columns not named correctly. At the moment the wizard requires two columns which must follow an exact naming scheme:

- The first column needs to be called "Sentiment" and must contain a sentiment or an empty value (for the rows for which you want a sentiment)
- The second column needs to be called "Text" and contain the text for each (missing) sentiment

We are aware that the error message and the strict naming requirement is suboptimal :-[

Regards,
Marco

lmcdowell

Hi Marco. Thank you for the reply. I am so disappointed because I have named the columns exactly as you described and they contain the data that you describe. Still, I am getting the error message every time I run the wizard. I so wanted this to work for our data set. Do you have any other suggestions?

Marco_Boeck

Hi,

could you provide me with a snippet of the data in a file that causes your problem (you can send me a PM here)? I may be able to figure out what is wrong (and create a ticket if it's a yet unknown bug).

Regards,
Marco