sentence_eval: Evaluate One Sentence

In simon-hans/CEDARS: Simple and Efficient Pipeline for Electronic Health Record Annotation

Evaluate One Sentence

Description

Assesses if one sentence matches the search query and marks positive tokens/concept unique identifiers (CUI's).

Usage

sentence_eval(
  unique_id,
  unique_sentences,
  annotations,
  query_vect,
  keyword_mask,
  keyword_elements,
  cui_mask,
  cui_elements,
  use_negation
)

Arguments

unique_id	Sentence ID, i.e. row number in annotations dataframe.
unique_sentences	Dataframe of unique sentences for the patient.
annotations	NLP annotation dataframe.
query_vect	Vector of search query components.
keyword_mask	Query vector with only keywords.
keyword_elements	Vector of search query keywords.
cui_mask	Query vector with only CUI's.
cui_elements	Vector of search query CUI's.
use_negation	Should negated items be ignored in the keyword/concept search?

Value

If sentence matches query, returns sentences with marked tokens/CUI's, otherwise NA.

simon-hans/CEDARS documentation built on Feb. 14, 2024, 3:16 a.m.