Abstract
Results of a content analysis of tobacco-related Twitter posts (tweets), with particular reference to e-cigarette and hookah related tweets. Descriptive results reveal more positive than negative sentiment toward tobacco among Twitter uses, as well as several apparent disconnects between public perception and tobacco-related health research, presenting opportunities for improved outreach and education. Additionally, machine classification of tobacco-related posts shows a promising edge over strictly keyword-based approaches, paving the way for automated tobacco surveillance applications.