Visual query tools for uncertain data in space and time

Katherine Mary Malan · Open University of Cape Town (University of Cape Town) · 2004

Uncertainty in data is a common problem.In this study we address particular challenges around the visual querying of data with vague locations in time and space.Our study is based on an archive of African art.In this archive there is a mixture of contemporary and traditional art.Many of the artworks have vague locations in time and space.By this we mean that, although there is some idea of when and where an artwork originated, the precise information is unknown.The aim of our project was to develop query and visualisation tools for this archive of African art.Our query system had to be designed to cope with different degrees of uncertainty in data, both in time and space.One of our initial ideas was to use a geographic map to drive the interface to the system.Items were plotted as a point on the map at some random location which fell within the given location of uncertainty for the item.However, user feedback indicated that this approach was problematic, as users assumed that the point on the map was the actual location of the artwork.A better approach was found in driving the interface around a time-line metaphor.Time is a one-dimensional attribute, making it possible to use the second dimension to display uncertainty information, while still retaining a two-dimensional interface.Objects are represented on the time-line as either a dot, if an exact time is known, or a line, if the time is uncertain (the length of the line depicts the length of the period in which the object may have been created).Results of user studies showed that the time-line based approach was clear to users.In addition to the display of uncertain data, we also developed techniques for storing and querying uncertain data.We show how data can be stored at different levels of uncertainty, without losing any available detail, and how queries can be expressed to include both certain and uncertain data.We use the concepts of dynamic queries; add to this a 2D query tool for performing spatial queries and tools for enabling Boolean combinations of queries.Results of prototype-based user tests indicated that users could successfully perform simple queries using the visual query tools.We have combined these techniques into a single visual query system.In this way, we show how it is possible for novice users to easily query large multimedia archives with complex uncertain attributes.6 Visual query system 6.1 User interface . . . . . . . . . . .6.1.1Probing.......... 6. 1.2 Selecting a subset of items

Read the paper · More papers on PaperTik