The Development of an Algorithm Using the TFIDF / TF Index toExtract Automatically the Set of Keywords of Corpus about Fields Related to Emergency Management

-A Case Study Utilizing Web News Articles for the 2004 Niigata-Ken-Chuetsu Earthquake Disaster-

Abstract

Based on the web news archive of a total of 2,623 articles highlighting the 2004 Niigata-ken-Chuetsu Earthquake disaster, we developed an algorithm to extract automatically the set of keywords which can characterize any time phase of disaster process from the start of an event to the termination of Emergency Operation Center. And we applied a visualization method with use of extracted keywords set and their feature quantify. Extracted keywords by the algorithm and the text mining result helps Cross Media Database (XMDB) users to search information and intelligence from corpus about fields related to emergency management.