Home > Geovisual Analytics > Paving the Way for Geovisual Analytics

Paving the Way for Geovisual Analytics

The human visual system can reliably identify trends in datasets using various channels of visualization. The success story of visualization in accomplishing core analytical tasks has made it a mainstay in various domains, including the Geographic Information Science (GIS). The core analytical tasks are data exploration, decision making, and predictive analysis. Geospatio-temporal data is ubiquitous, and as the data grows in complexity and size; analytical tasks become more complex too. In the times of geospatial big data, geovisual analytics has begun to gain traction. In this paper, we give a brief overview of geovisualizations, discuss geovisual analytics through a case study, and list out some of the persistent challenges in geovisual representation and analysis.

 

1. Introduction

 

There is a frequently used phrase, which says that “80% of all information contains some geo-reference.” Hahmann and Burghardt [10] have reformulated this phrase using evidence to a “60% assertion.” Nonetheless, the geo-referenced data remains to be relatively more ubiquitous than non-geo-referenced data. Consider the NISAR (NASAISRO Synthetic Aperture Radar) mission [30], which the two space organizations are going to jointly launch in 2021. The mission is expected to generate around 85 TB of data daily [6]. The radar imaging satellite will be designed to use dual frequency for remote sensing for studying natural processes and estimating earth observations (e.g. biomass, surface deformation, soil moisture, etc.). The mission is estimated to gather about 140 PB in its 3-year mission.
Overall, the message is loud and clear that geospatial big data is going to stay and grow, more so. Hence, its analytics cannot be ignored. Li et al. [20] have discussed how visualization provides a class of methodologies to gain human insight to geospatial big data. Visualizations provide yet another channel for identifying outliers, building and verifying hypotheses, and most importantly, “seeing” dominant trends and patterns. Visualization as a domain has contributed towards techniques for summarization and exploration of data, as well as “computational steering”. Computational steering [29, 28] is the process of improving on computations by using intermediate data visualizations, which effectively influence the workflow. While computational steering is used in computational simulations, a more recent trend has been observed in the form of progressive visual analytics.

1.1 What is Visual Analytics and How is it Different from Visualizations?

 

Keim et al. [17] have defined visual analytics as a combination of conventional data mining and interactive  visualizations in a data analytic workflow, for understanding, inferences, and decision-making support for big data. In all interactive applications today, visualizations have paved the way for visual analytics. This evolution has happened because a stand-alone visualization or a group of visualizations carefully assembled together is no longer sufficient for making sense of the big data. Visual analytics processes, as shown in Figure 1, could be non-deterministic, as well, and form a feedback loop which terminates only when “visualization requirements” are met.

 

Fig. 1: The visual analytics workflow, as defined by [17], has been modified to include (a) streaming data; and (b) data feed from data mining. Use of streaming data leads to incremental visual analytics [31] and use of data in (b) leads to progressive visual analytics [33, 4].
A more recent concept, “progressive visual analytics,” [33], entails the use of visualization to study the “meaningful partial results” given out during the execution of data analytic workflow. Fekete and Primet [8] have proposed progressive analytics to be a synonym for “progressive computations for data analytics,” with consideration for low latency. Including visualizations to support progressive analytics, Badam et al. [4] have differentiated between progressive and incremental visual analytics – where the latter caters to data streaming [31]. Their prototype for “progressive visual analytics” combines both incremental as well as progressive visualization with algorithmic steering, i.e. iterative control over execution of a computational process [28].

 

1.2 Geovisualizations and Geovisual Analytics

For visual analysis, specifically, in geospatial applications, MacEachren and Kraak have introduced geovisualization [26] as an amalgamation of the areas of visualization, scientific computing, and geographic information systems (GIS) to make sense of geospatiotemporal data. MacEachren et al. [25] have defined the four functions of geovisualizations to be “explore, analyze, synthesize, and present.” They also have defined the space of  geovisualizations to be a cube with axes for task types, user types, and user interaction level.

In 2007, Andrienko et al. [1] have introduced the term geovisual analytics, setting the stage for its widespread usage as well as development. They have pushed forward the research agenda for “geovisual analytics for spatial decision support systems” and differentiated geovisual analytics from the regular visual analytics using three characteristics, namely, complex spatio-temporal nature of the data involved, multiple stakeholders, and tacit criteria and knowledge. They have illustrated the application of geovisual analytics in the example of emergency response in a disaster-affected region. Specifically, to plan the evacuation of people from the region, the planner needs information of the geography of the place, the road network, the hospital facilities en route, high-risk personnel, etc. We see that the problem blows out to be a multi-criteria optimization one. The decision making of the planner, in this case, may be alleviated by making inferences from different “slices” of the data. Visual analytics can also be used to study the dynamism in this example, say, we can progressively refine the solution using both the incoming streaming data of the evacuation process as well as partial results from other data mining process (Figure 1).

In this paper, we sample some of the popularly used geovisualizations in Sections 2 and 3, demonstrate geovisual analytics using a case study in Section 4, and finally articulate the research challenges in geovisual analytics. Geo-referencing makes the use of cartographic products an important element of the visualizations. Hence, we discuss map-based visualizations in Section 2, and other related visualizations in Section 3. The geovisualization techniques described in this paper are not exhaustive, however. For instance, we have not covered the role of information visualization techniques as well as user interactions in all of these visualizations. For details of these topics in the context of geovisualization, they can be found in parts in [2, 3, 12]. In the current article, we focus on discussing popularly used geovisualizations and move on to show how they influence geovisual analytics in a specific case study.

Leave a Comment:

Your email address will not be published. Required fields are marked *