search

stanfordnlp / stanza

Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages
https://stanfordnlp.github.io/stanza/
Other
7.25k stars 888 forks source link

Stanza 1.8.1 mislabels `nsubj` relationship as `compound` #1363

Open khannan-livefront opened 7 months ago

khannan-livefront commented 7 months ago

Describe the bug I came across a set of sentences where Stanza mistakenly labels nsubj relationships as compound. This occurs with a noun in relation to the word "hops".

To Reproduce Steps to reproduce the behavior:

  1. Go to http://stanza.run/ or input into stanza either sentence:
Zac hops, and Rod is on the box. 

The box hops and hops.
  1. See error – stanza fails to find the nsubj between "Zac hops" and "box hops":

Screenshot 2024-03-06 at 1 12 53 PM

Expected behavior The parse returns a nsubj relationship between "Zac hops" and "box hops".

Environment (please complete the following information):

Additional context This issue also appears in Stanza 1.8.1. Have not tested it with Stanza 1.7.x. Screenshot is from Stanza 1.6.1.