Understanding and predicting social behaviour using digital trace data – Supporting public policy decisions with real-time online listening-based social-, linguistic-, and simulation models

The main objective of the project is to create a policy-decision support platform that is able to monitor in real-time the opinion and reaction of the Hungarian population on policy topics, using textual data generated in the online space.

At present, policy-making relies mostly on traditional quantitative data collection methods (survey) and partly on qualitative research (focus groups). Compared to traditional methods, data generated in the digital space has several advantages. The large amount of data provides an opportunity for high-granularity analysis of long periods, which provides an opportunity to study the impact of various interventions and to monitor the impact of external conditions. A further advantage of analyzing the discourse generated in online space is that this type of data collection analyse people’s communication without intervention, and observed opinions are not influenced by research design.

By the end of the project, we plan to create a platform that continuously collects online opinions on specific policy topics, sorts them and can easily generates reports using this data along pre-defined criteria.

Keywords: public policy, online listening, text mining, NLP