hi.
thanks to my teacher I've entered the fantastic world of Rapidminer. I love it, even though Im still a newbie.
Im trying to proceed with a text classification modeling starting with an Excel file with two columns:
Column1 Column2
ROW 1 attribute (text) Label(binomial: simply 0 for negative review and 1 for positive review)
up till now we only work with positive reviews in Txt stored in a folder and negative reviews in Txt stored in another folder, we defined the two of them as positive class and negative class.
I've tried to proceed like this with Read Excel - Process Documents (Tokenize, remove stopwords and case) - Validation (training with SVM + Applay model and Performance)
I've used Nominal to numerical to avoid SVM capacity problems, but as a result I get only the rooted mean square error, in the Performance vector.
I was looking for the Accuracy of my model instead... sorry for the bad question, I hope somenody can help.
Can I use a txt file as an alternative? see attached file.
thanks a lot in advance