Developing a framework for an advisory message board for female victims after disasters: A case study after east Japan great earthquake

摘要

After the sudden occurrence of East Japan Great Earthquake on 11 March 2011, triple disasters crippled the regular life of citizens of East Japan. A lot of people were affected, especially women victims suffered from different problems and worries: they had to care for elders, raise children, and find new jobs. Women also had specific needs of commodities for everyday life. Administrative authorities wanted to recognize women victims' specific problems and provide them appropriate supports. However, it was difficult to grasp women victims' requirements properly, because they were really patient and their needs were sometimes neglected under the environmental pressure. Conducting interviews and taking questionnaire from women victims is one way to gauge their needs, but it is time-consuming and labor intensive. This work proposes a framework for the development of an advisory message board for women victims on the web in which women victims can post their messages freely. The computational technologies are used here for analyzing digital media data in order to improve the lives of underserved or underprivileged people in case of disasters like earthquake. Text mining technologies are developed for automatic analysis of the messages to find out the specific needs of the victims and the change of needs with time and support them with proper advice. The proposed method uses latent semantic analysis (LSA) to extract the hidden topics and change of topics over time. As a case study, text messages from several on-line social media over a period of 2 years after the East Japan Great earthquake are collected and analyzed. It has been found that LSA-based technique is more effective in extracting the change of needs over time than graph-based topic extraction method. The final aim of this work is to develop the framework of advisory message board for women which will help the authority to find out the special needs of women victims after any future disaster and support them.