Mixing text mining analysis and agent based modelling methodologies.: A case study to analyze a controversy

Résumé 0

This paper starts from methodological issues dealing with sociological and quantitative interpretation of qualitative and discontinued data when analyzing controversies from a large press corpus. Then authors offer a new approach mixing text mining analysis and agent based modeling. The study case dealing with the controversy of abnormal disappearance of honey bees (\textit{apis mellifera}) among French speaking journalistic during 13 years is mobilized to describe the different steps of this heuristic framework. First articles are tagged with three stances to report the problematic phenomenon, a uni-factor cause, i.e. the use of pesticides, a multi-factor cause, i.e. including one other factor different than pesticides at least, or the absence of understanding. Second, variations of the proportions of agents explaining the issue either with unifactor or multifactor causes are obtained with modeling. Assuming agents follow dispositional or positional social influence in their interactions to report the facts, their associated networks are extracted from the data applying a network randomized model of opinion dynamics. Third, from those distributions the possible topology of actor networks can be questioned back with others qualitative methods, either ethnographic or interviews.