WEB-BASED VISUAL EXPLORATION AND ERROR DETECTION IN LARGE DATA SETS: ANTARCTIC ICEBERG TRACKING DATA AS A CASE

• Main Authors: Connie A. Blok, Ulanbek Turdukulov, Barend Köbben, Juan Luis, Calle Pomares
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Antarctic; Antarc*; Iceberg*
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.501.7363; http://kartoweb.itc.nl/kobben/publications/2009/blok_et_al_ica2009-icebergs.pdf

Polar iceberg data are – amongst others – used by scientists who are interested in complex phenomena like global warming, climate change and changing habitats. Other main users of the data are organisations and individuals involved in navigation and engineering that need to take iceberg positions and behaviour into account for their activities. Iceberg positions are identified, tracked and, together with some other attributes, made available on the Web. Data tracking and storage are manually performed, and therefore error-prone. If these errors are not detected and handled, they will propagate and negatively influence the quality of the derived information. The main objective of this paper is to demonstrate that an interactive environment to visually explore the characteristics of spatio-temporal data can also be used to detect errors in the data set. We report on the method used to design and implement a web-based environment to visually explore iceberg characteristics. Although the environment was not intended for error detection, we found that several errors in the data could be quickly discovered. Main limitations in the current propotype are described, together with recommendations for improvements.