Visual exploration of temporal object databases
Chaouki Daassi, Marlon Dumas, Marie-Christine Fauvet, Laurence Nigay, Pierre-Claude Scholl · 2000
Two complementary families of users' tasks may be identified during database visualization: data browsing and data analysis. On the one hand, data browsing involves extensively exploring a subset of the database using navigational interaction techniques. Classical object database browsers provide means for navigating within a collection of objects and amongst objects by way of their relationships. In temporal object databases, these techniques are not sufficient to adequately support time-related tasks, such as studying a snapshot of a collection of objects at a given instant, or detecting changes within temporal attributes and relationships. Visual data analysis on the other hand, is dedicated to the extraction of valuable knowledge by exploiting the human visual perception capabilities. In temporal databases, examples of data analysis tasks include observing the layout of a history, detecting regularities and trends, and comparing the evolution of the values taken by two or more histories. In this paper, we identify several users' tasks related to temporal database exploration, and we propose three novel visualization techniques addressing them. The first of them is dedicated to temporal object browsing, while the two others are oriented towards the analysis of quantitative histories. All three techniques are shown to satisfy several ergonomic properties. Keywords: temporal database, object database, data browsing and analysis, visualization technique. 1