Many researches have been conducted extracting keyword relationships using keyword hit counts of a search engine. Since the relationships vary with multiple viewpoints, it is important to recognize transitions between them. However, previous methods are not useful for variance recognition. This paper proposes a visualization method for transition recognition between keyword relationships. The interface shows animations where the differences move between two relationships in order to help users recognizing the transitions. The proposed method shows the relationships as maps including islands, houses, bridges, rail tracks, and roads in order to support users in recognizing the whole relationships. Experimental results showed that the interface supported users in understanding the whole transitions between maps.

View full abstract