how can I annotate my text to argumentation components?

Dear all,

I hope you are doing well If you remember I have trained custom NER using prodigy. I am really happy with the model

here it is an example:

basically I have this kind of entities:

LONG,DATE,TIME,NAMES,STAR,PLAN,GEOM,...

then using these entities extracted explicit observational sentences (I can call them evidence in the context of argument mining)

you can see here

like those sentences , now I am looking for a way to extract "claims"

how can I do it using prodigy?

do you think the same approach works?

generally speaking what would be your recommendation to annotate using prodigy my corpus for the aim

"argumentation mining/extraction" or work on the structure of argumentation?

many thanks