Abstract
This paper proposes an interactive context search system for exploratory analysis of exchange rate data. As the fluctuation of exchange rate is influenced by various kinds of events, it is important to know how a certain type of events will influence the fluctuation, and type of events common to similar time periods. This paper defines two types of context search functions: summarizing events occurred in similar time periods, and searching events along with corresponding time periods. In order to find similar time periods, this paper employs the approach of Kansei Engineering. That is, suitable similarity calculations for temporal data is selected according to the level of granularity a user is interested in. This paper shows the experimental results on similarity calculation with different granularities, and those with test participants conducting exploratory analysis. Experimental results show the participants could analyze relationship between exchange rate and events with using the proposed system.
