An evaluation system for news video streams and blogs

Abstract

When we want information on current events, we often view news programs on TV or news streams on Web sites. A news video stream consists of several scenes, and viewers often gain a broad understanding of the news by viewing scenes in the given order. Since a viewer's opinion of a news topic will depend on the scene order, we have developed a method for extracting blog entries expressing a particular point of view regarding news topics where we use a form of evaluation and categorization similar to that based on news scene order. In this paper, we describe our method of news evaluation based on news scene order and explain how a blog search can be done using a news stream.