Yes/no annotations with textcat.manual

, ,
1

Hello all,
I am new to Prodigy, and I am trying to work out how to classify a text as yes/no according to two categories, e.g. is the text funny? is the text about computers? Is this possible? Thank you for your help. I am using textcat.manual.

2

Hi Joao,

You have to use textcat with one label for this case.

for example,
I am using the following text as sample input,

image
image749×103 16.5 KB

textcat manual with one label-

image
image699×22 2.74 KB

Here "FUNNY_OR_NOT" will be you annotated dataset and "IS_FUNNY" is the only label.

It will initiate the tool and you can annotate the text as accept/reject or ignore.

image
image1296×488 180 KB

and, your result will look like this,

image
image1392×34 8.15 KB

cheers.

1 Like
3

Thank you. What if, for your example, you also wish to classify, in addition, say 'Is this comment about cats?'. Thank you for your help.

4

checkout following link for multi-label categorization,