Skip to main content

Table 2 Example labeling functions for the MetaCyc corpus

From: Extracting chemical reactions from text using Snorkel

Example labeling function

Description

LF_keyword_context

If there is a word of a given list, such as reduce, oxidize, transform, or afford between the two words, we label TRUE

LF_sep_verb

If the chemicals are separated by a verb, we label TRUE

LF_argument_order

If the candidate product is before the candidate substrate, we label FALSE

LF_followed_ase

If one of the chemicals is followed by a word that ends with “ase”, we label FALSE

LF_sep_or

If the chemicals are separated by the word or, we label FALSE