Abstract
When a natural disaster occurs, victims’ information needs vary with their personal attributes. (e.g. age, gender, occupation and so on), and the needs also vary with the time passed from the disaster. Thus, it is required to make a system that enables victims to easily satisfy their information needs considering their personal attributes and time phase. As a fundamental study before making such a system, we did time series analysis of circumstances of victims by applying text mining method for newspaper articles about the 2011 Great East Japan Earthquake. As a result, we clarified that it is possible to find the time series variation of correlation between some major personal attributes and provided information from newspaper articles.
