KnoVA: A Reference Architecture for Knowledge-based Visual Analytics

The considerably increasing Internet-based information processing leads to a rapidly increasing growth of the available data. Automatic analysis approaches are no longer sufficient for the analysis of this data. One approach to overcome this information overload phenomenon is visual analytics (VA). It combines the strengths of computers to process large amounts of data with human strengths such as flexibility, intuition, and contextual knowledge. In the process of VA experts apply knowledge. In many settings they apply similar knowledge continuously in several iterations across various tasks. Therefore a demand for concepts and methods to prevent needless repetitive analysis steps can be identified. This thesis presents KnoVA, a reference-architecture for knowledge-based visual analytics systems and evaluates it in two applications domains, automotive and healthcare. KnoVA consists of four parts: a model of the analysis process, a metadata model for knowledge-based visual analytics systems as well as concepts and algorithms for the extraction and the re-application of knowledge.

Level thesis:degree_level

thesis.doctoral

Grantor dc:publisher

Universität Oldenburg

Year

2012

Author dc:creator

• Flöring, Stefan

Contributors dc:contributor

• Appelrath, Hans-Jürgen
• Isenberg, Tobias

• Wissensbasiertes System, Visualisierung, Analyse, Wissensextraktion, Informationsüberlastung

Repository record source_url

http://oops.uni-oldenburg.de/1450

OAI identifier oai:identifier

oai:oops.uni-oldenburg.de:1450