Skip to main content

Table 3 Keywords of the NLP algorithm for the extraction of smoking status in the i2b2 2006 shared task

From: A clinical text classification paradigm using weak supervision and deep representation

Current Smoker

(does|has|continues to) smoked?, uses tobacco, active smoker, (current|currently) (smoker|smoking), current smoker, tobacco use\W*(yes|still using|still smoking|smokes)

Past Smoker

(stop|stopped|quit|quitted|discontinued) (tobacco|smoking), (previous|prior|remote|distant|former|ex-|ex) (tobacco|smoker), stop(ped)? smoking, tobacco use\W*(smoked|quit), smoking\W*(used|former)