Created
April 27, 2020 15:49
-
-
Save pafonta/8232f7c8405ca8800aefbfff2b160f5f to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
nlp2 = load_model(MODEL) | |
nlp2.add_pipe(nlp2.create_pipe("merge_entities")) | |
nlp2.add_pipe(nlp2.create_pipe("merge_noun_chunks")) | |
def extract_relations_umls(text, umls): | |
ents = {x[0].lower(): x[2] for x in umls} | |
rels = [] | |
for x in nlp2(text): | |
if x.dep_ != "ROOT": | |
if x.ent_type_ == "ENTITY": | |
if x.head.ent_type_ == "ENTITY": | |
rels.append((x.head.lower_, x.dep_, x.lower_)) | |
else: | |
if x.head.ent_type_ == "ENTITY": | |
children = list(x.children) | |
if children and children[0].ent_type_ == "ENTITY": | |
rels.append((x.head.lower_, x.dep_, x.lower_)) | |
rels.append((x.lower_, children[0].dep_, children[0].lower_)) | |
return [(ents.get(x, x), y, ents.get(z, z)) for x, y, z in rels] | |
rels = extract_relations_umls(text, umls) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment