東日本大震災について報じられたウェブニュースコーパスの基礎的解析

抄録

The web information, especially mass media and social media information, have a more increasing role in create operational picture about a crisis since the Great East Japan earthquake and tsunami disaster. In this paper, we have developed and basically analyze web news corpus of the earthquake and tsunami disaster with text mining method intended to clear the whole picture. The corpus consists of over 60,000 web news articles which tagged its topic name. Sub corpora about Nuclear power plant accident, Iwate, Miyagi, Fukushima, Aomori, Ibaraki, and Chiba from original corpus, and analyzed by comparison of extracted keywords and their chronological behaviors.