Urška Demšar

Space, time and visual analytics

Visual analytics aims to combine the strengths of human and electronic data processing. Visualisation, whereby humans and computers cooperate through graphics, is the means through which this is achieved. Seamless and sophisticated synergies are required for analysing spatio-temporal data and solving spatio-temporal problems. In modern society, spatio-temporal analysis is not solely the business of professional analysts. Many citizens need or would be interested in undertaking analysis of information in time and space. Researchers should find approaches to deal with the complexities of the current data and problems and find ways to make analytical tools accessible and usable for the broad community of potential users to support spatio-temporal thinking and contribute to solving a large range of problems.

Space-time density of trajectories: exploring spatio-temporal patterns in movement data

Modern positioning and identification technologies enable tracking of almost any type of moving object. A remarkable amount of new trajectory data is thus available for the analysis of various phenomena. In cartography, a typical way to visualise and explore such data is to use a space–time cube, where trajectories are shown as 3D polylines through space and time. With increasingly large movement datasets becoming available, this type of display quickly becomes cluttered and unclear. In this article, we introduce the concept of 3D space–time density of trajectories to solve the problem of cluttering in the space–time cube. The space–time density is a generalisation of standard 2D kernel density around 2D point data into 3D density around 3D polyline data (i.e. trajectories). We present the algorithm for space–time density, test it on simulated data, show some basic visualisations of the resulting density volume and observe particular types of spatio-temporal patterns in the density that are specific to trajectory data. We also present an application to real-time movement data, that is, vessel movement trajectories acquired using the Automatic Identification System (AIS) equipment on ships in the Gulf of Finland. Finally, we consider the wider ramifications to spatial analysis of using this novel type of spatio-temporal visualisation.

Identifying Critical Locations in a Spatial Network with Graph Theory

Effective management of infrastructural networks in the case of a crisis requires a prior analysis of the vulnerability of spatial networks and identification of critical locations where an interdiction would cause damage and disruption. This article presents a mathematical method for modelling the vulnerability risk of network elements which can be used for identification of critical locations in a spatial network. The method combines dual graph modelling with connectivity analysis and topological measures and has been tested on the street network of the Helsinki Metropolitan Area in Finland. Based on the results of this test the vulnerability risk of the network elements was experimentally defined. Further developments are currently under consideration for eventually developing a risk model not only for one but for a group of co-located spatial networks.

Investigating visual exploration of geospatial data: An exploratory usability experiment for visual data mining

This study presents a small exploratory usability experiment with the goal to observe how people visually explore geospatial data. The well-known iris dataset from pattern recognition was put into geographical context for this experiment, in order to provide the participants with a dataset with easily observable spatial and other relationships. The participants were given free hand to explore this dataset with a visual data mining system in any way they liked. The protocols collected during the experiment with the thinking-aloud method were analysed with the aim to understand what types of hypotheses the participants formed, which visualisations they used to either derive, confirm or reject their hypotheses and what exploration strategies they adopted.

Combining Geovisual Analytics with Spatial Statistics: the Example of Geographically Weighted Regression

Abstract An attempt is made to facilitate interpretation of the results of a spatial statistical method – Geographically Weighted Regression (GWR) – using a geovisual exploratory approach. The GWR parameter space is treated as a multivariate dataset and explored in a geovisual exploratory environment with the goal to identify spatial and multivariate patterns that describe the spatial variability of the parameters and underlying spatial processes.

Knowledge Discovery in the Environmental Sciences: Visual and Automatic Data Mining for Radon Problems in Groundwater

Efficiently exploring a large dataset with the aim of forming a hypothesis is one of the main challenges in environmental research. The exploration of georeferenced environmental data is usually pe ...

Visual Comparison of Moving-Window Kriging Models

Abstract Kriging is a spatial prediction method that is core to the geo-statistical paradigm. Commonly applied in the environmental sciences, it enables a prediction at an unsampled location coupled with a measure of confidence in its accuracy. Many variations of kriging exist, some of them complex, especially those that allow many parameters to vary spatially. Calibrating such a kriging model and interpreting its results can therefore be quite daunting. We suggest that visualization and visual exploration can help with this task. In particular, we focus on the moving-window kriging model, evaluating three newly developed robust variants of this model against a basic counterpart. We use star icon maps and plots to visually explore model results to evaluate model parameterization, specification, and performance.