Skip to Main Content
Table 1

Features used by the frame classifier. Example values for the annotated graph in Figure 1 are given in parentheses.

Feature
Type
Description and example value
target_lemma atomic lemma of the target node (blink
frames set frames that can be evoked by the target verb ({Body_movement}) 
voice binary voice of the target node (active) 
parent_word set lemma of the parents of the target node ({and}) 
parent_POS set part of speech of the parents of the target node ({CC}) 
rel_to_parent set grammatical relations between the target node and its parents ({conj}) 
parent_has_obj binary whether any parents have an outgoing “object” relation (no) 
dsubcat atomic subcategorization frame, the multi-set of all outgoing relations of the target node (dobj
child_word_set set lemma of the children of the target node ({eye}) 
child_dep_set set outgoing relations of the target node ({dobj}) 
child_word_dep_set set pair (lemma, relation) for the children of the target node ({(eye, dobj)}) 
Feature
Type
Description and example value
target_lemma atomic lemma of the target node (blink
frames set frames that can be evoked by the target verb ({Body_movement}) 
voice binary voice of the target node (active) 
parent_word set lemma of the parents of the target node ({and}) 
parent_POS set part of speech of the parents of the target node ({CC}) 
rel_to_parent set grammatical relations between the target node and its parents ({conj}) 
parent_has_obj binary whether any parents have an outgoing “object” relation (no) 
dsubcat atomic subcategorization frame, the multi-set of all outgoing relations of the target node (dobj
child_word_set set lemma of the children of the target node ({eye}) 
child_dep_set set outgoing relations of the target node ({dobj}) 
child_word_dep_set set pair (lemma, relation) for the children of the target node ({(eye, dobj)}) 
Close Modal

or Create an Account

Close Modal
Close Modal