Increasing Relevance of Smoking Cessation Messages in an Online Software Agent Environment

Abstract

An online software agent that helps smokers quit was designed and tested. We created a library of categorized smoking cessation messages using meta-data corresponding to the Stages of Change Theory. A feedback process was developed that used individual participant’s relevance ratings and a message similarity search algorithm. A pilot study of university students who smoke or had recently quit was performed. Participants were randomly assigned to one of three groups: one received generic, non-tailored messages; another received tailored messages based on their answers to questions about their smoking and quitting behavior; and another received messages selected through tailoring and feedback. In the feedback-driven group, participants reported relevance of the messages received averaged higher than the other two groups. There was also a highly significant correlation in this group between relevance and social presence, which indicates the "feeling" of interacting in an interpersonal manner.