Supporting Consecutive Data Exploration by Visualizing Spatio-temporal Trend Information

This paper reports on a study in which a proposed interface that visualizes spatio-temporal trend information for exploratory data analysis was realized. The proposed system supports users' analysis of time-series data in an exploratory manner. Such exploratory analysis requires repeated access to various types of information related to the users' interests such as texts and numerical data. During analysis of time-series data, it is essential that users understand temporal changes and the cause of those changes. The proposed system helps users by facilitating understanding of the temporal changes in time-series data and the causes of those changes. The implemented interface currently utilizes newspaper articles, maps, and statistical data as data modalities.