Disaster Anxiety Measurement and Corpus-Based Content Analysis of Crisis Communication

The aim of this study is to develop a methodology for evaluating public anxiety arising from disaster events and to clarify the nature of crisis communication between the government and citizens. Preparing for catastrophes that may happen in the future is an important issue in risk management. In the Great East Japan Earthquake, Twitter was used widely as a means of sharing information about the disaster. This paper proposes a methodology for measuring anxiety by determining semantic orientations of risk assessments and clarifying the contents and structures of crisis communication systems by referencing the corpus of government announcements and Twitter during the week after the Great East Japan Earthquake. Objectivity of this paper has been ensured by applying natural language processing and text mining techniques based on corpus linguistics.