CAREER: Visual Analytics Algorithms for Spatiotemporal Analysis Traditionally, visualization is one of the most important, and commonly used, methods of generating insight into large scale data. Particularly for spatial data, the translation of such data into a visual form allows users to quickly see patterns, explore summaries and relate domain knowledge about underlying geographical phenomena that would not be apparent in tabular form. However, several critical challenges arise when visualizing and exploring these large spatiotemporal datasets. While, the underlying geographical component of the data lends itself well to univariate visualization in the form of traditional cartographic representations (e.g., choropleth, isopleth, dasymetric maps), as the data becomes multivariate, cartographic representations become more complex. Multivariate color maps, textures, small multiples and 3D views have been employed as means of increasing the amount of information that can be conveyed when plotting spatial data to a map. However, each of these methods has their own limitations. Multivariate color maps and textures result in cognitive overload where much time is spent trying to separate data elements in the visual channel. In 3D, occlusion and clutter remain fundamental challenges for effective visual data understanding. Utilizing small multiples can help in side-by-side comparison, but their scalability is limited by the available screen space and the cognitive overhead associated with pairwise comparisons. Such problems are further compounded when a temporal aspect is added to the data. Such data is often explored using animation and controlled playbacks; however, this forces end users to rely on their memory and do sequential comparisons between animation frames. As the number of spatial units being explored increases, this quickly becomes intractable. Such challenges call for a transformative view to explore large spatial data. While intensive efforts have been spent on developing better algorithms and techniques for data management, querying and analysis of such data, much less attention has been paid to the design of effective visual analytics solutions. Instead of being confined to the original spatiotemporal domain, this proposal seeks to both extend traditional visual representations and develop novel views for showing correlations and clusters. Underlying these novel views is also the need for visual representations in which the manipulation of the representation is directly tied to the underlying computational analytics process. By rethinking the space of interactions and the role of the user, the broad, long-term goal of this CAREER proposal is to develop a suite of techniques that will support the effective exploration of large scale spatiotemporal data. The proposed framework will include the following: Analytical Brushing: a palette of analytical brushes where users can quickly generate cluster queries and similarity searches; Parallel Matrix Sets: local correlations result in spatial clusters, with properties and members that change at each time step, these correlations and membership changes can be represented as a series of correlation matrices over sets of time; Point Process Flow Vis: point process data can be transformed into volumetric data from higher order derivatives can be used to visualize spatiotemporal flow; and The Visual Recommender: by capturing historical interactions of a variety of users, underlying statistical matching algorithms can be applied to suggest potential views of interest at varying levels of workflow interruption states. These four components provide both novel visual mappings that abstract data relationships at varying levels of detail and serve as a novel navigation interface for visual analytics.
|Effective start/end date||8/1/14 → 7/31/20|
- National Science Foundation (NSF): $466,764.00