Competence Modeling in Twitter: Mapping Theory to Practice

Availability of “big data” from the Social Web provides a unique opportunity for synergy between the computational and social sciences. On one hand, psychologists and social scientists have developed and established models of human competence, credibility, trust and skill over many years. Currently, much research is being conducted by computer scientists to evaluate these human-behavioral aspects using real-world data from Twitter and other sources. However, many of these algorithms are formulated in an ad-hoc way, without much reference to established theory from the existing literature. This paper presents a framework for mapping existing models of human competence and skill onto a real world streaming data from a social network. An example mapping is described using the Dreyfus model of skill acquisition, and an analysis and discussion of resulting feature distributions is presented on four topic-specific data collections from Twitter, including one on the 2014 Winter Olympics in Sochi, Russia. The mapping is evaluated using human assessments of competence through a crowd sourced study of 150 participants.