Event mining from the Blogosphere using topic words

Yoshihiko Suhara, Hiroyuki Toda and Akito Sakurai

In this paper, we propose a method of extracting "action relations" between related topic words from Japanese weblogs (blogs). An action relation is a tuple of agent, target and predicate. Our method obtains blog articles that contain two keywords by AND search and outputs action relations constructed from the predicate and the two keywords with case particles that follow. However, since words are often omitted after they appear once, sentences with both of the keywords are scarce. Our method overcomes this problem by combining sentences that contain the common predicate and locate nearby.

Poster

Available as PDF