Default controlled-vocabulary lexicons for the lexicon feature block

The small, shippable lexicons blk_lexicon scores each column against, including the public-domain person/place gazetteers (first/last name, US city) that are the learned router's strongest features. Each named entry becomes lex_<name>_value and lex_<name>_token rate features. Extend by passing your own named list of character vectors to fx_extract_features().

Usage

default_lexicons()

Value

A named list of lowercase character vectors.

See also

fx_extract_features()