search

HazyResearch / flyingsquid

More interactive weak supervision with FlyingSquid
Apache License 2.0
314 stars 22 forks source link

Flying Squid for NER #15

Open david-waterworth opened 3 years ago

david-waterworth commented 3 years ago

I'm wondering how you'd go about specifying a dependency structure graph for a multi-label NER type problem. For example, the case where one label function tags "George Washington" as ["I-per", "I-per"] and another as ["ABS", "I-loc"].

At first glance, it seems like you'd define each task as assigning a single label to a single token, i.e.

Y1=George, Y2=Washington

I'm trying to figure out specifically how this would map to Figure 2 in your paper. Is it possible to specify the dependency graph such that the overlapping labels are correctly resolved (i.e. we don't end up with ["I-per", "I-loc"])?

ruanchaves commented 2 years ago

Hi @david-waterworth . Were you able to figure this out on your own? I'm also interested in using Flying Squid for NLP tasks.