The exploration of large geospatial data for finding patterns and understanding underlying processes is one of the challenges in geovisualization research. New methods are needed for effective extraction of patterns and appropriate visualization tools are necessary to support knowledge construction throughout the exploration process.
Based on an approach to combine visual and computational methods, a visualization environment has been developed to support visual data mining and knowledge discovery tasks. This environment integrates non-geographic information spaces with maps and other graphics that allow users to explore patterns and attribute relationships.
The development of the tool intends to facilitate knowledge construction using a number of steps that underline data mining and knowledge discovery methodology. In order to investigate the effectiveness of the design concept, an empirical usability testing is planed to assess the tool’s ability to meet user performance and satisfaction. In this test, different options of map-based and interactive visualizations of the output of a Self-Organizing Map (SOM) are used to explore a socio-demographic dataset.
The study emphasizes the knowledge discovery process based on exploratory tasks and visualization operations. This paper describes the usability framework used to guide the design, and examines key aspects of the evaluation of such visual-computational environment.
Source: The Pennsylvania State University
Author: Etien L. Koua | Menno-Jan Kraak