After the East Japan Great Earthquake, women victims have been suffering from many problems and worries: they are caring of elders, raising children, finding jobs, needs for female-specific items and so on. Administrative authorities want to recognize these women sufferers' specific problems and give them appropriate and timely advisory information. However, in general, it is difficult to grasp their needs timely because their environments and conditions change from moment to moment, and take a lot of labor for interviews or questionnaire investigations. This paper aimed to develop the data mining engine which analyzes the victims' requirements. The data mining engine analyzes and detects the current topics from the sufferers messages posted on social media societies. With data analyzing results, the system can detect victims' needs transitions.
