Designing Visual Analytics Methods for Massive Collections of Movement Data

Exploration and analysis of large data sets cannot be carried out using purely visual means but require the involvement of database technologies, computerized data processing, and computational analysis methods. An appropriate combination of these technologies and methods with visualization may facilitate synergetic work of computer and human whereby the unique capabilities of each “partner” can be utilized. We suggest a systematic approach to defining what methods and techniques, and what ways of linking them, can appropriately support such a work. The main idea is that software tools prepare and visualize the data so that the human analyst can detect various types of patterns by looking at the visual displays. To facilitate the detection of patterns, we must understand what types of patterns may exist in the data (or, more exactly, in the underlying phenomenon). This study focuses on data describing movements of multiple discrete entities that change their positions in space while preserving their integrity and identity. We define the possible types of patterns in such movement data on the basis of an abstract model of the data as a mathematical function that maps entities and times onto spatial positions. Then, we look for data transformations, computations, and visualization techniques that can facilitate the detection of these types of patterns and are suitable for very large data sets – possibly too large for a computer's memory. Under such constraints, visualization is applied to data that have previously been aggregated and generalized by means of database operations and/or computational techniques.