Visualisation

Advisor
Abstract

The work presents methods to support maintainers and developers of web applications in the field of software engineering and business management with visualization of web usage data. Usage data from an exemplary web application is analyzed by using mostly existing and also newly developed visualization and interaction methods.

The general concept of control circuits is combined with the basic information flow in and around web applications. The resulting model is presented as the Web Information Cycle (WIC). By that we present an exhaustive model for the information flow trough and around web applications. Within the WIC we present the positions of the data generating process for the analyzed usage data as well as the point where the results of the analysis provide a benefit for clients. For an adequate way of the data representation we present a selected set of existing visual and interaction methods. We put special focus on visual methods with a statistical background and describe the mosaic plot with its capabilities for the work. Moreover, we present a set of existing interaction methods and discuss their usefulness for the work. For a single task we discuss an individual approach and present a visualization method named Logsnakes. The method is implemented in R code for further use. A proposed data model along with the visualization and interaction methods is furthermore described by UML diagrams. Those are the basis for interviews with experts where the concept is evaluated with respect to its possible application in practice. The results from the interviews are presented and discussed critically.
Year of Publication
2008
Secondary Title
Institute of Visual Computing and Human-Centered Technology
Paper
Number of Pages
102
reposiTUm Handle
20.500.12708/11837
Publisher
TU Wien
Place Published
Vienna
S. Schnabl, “Investigating Web usage data with visual and explorative analysis methods”, Institute of Visual Computing and Human-Centered Technology. TU Wien, Vienna, p. 102, 2008.
Master Thesis
AC05036487
M. Bögl et al., “Cycle Plot Revisited: Multivariate Outlier Detection Using a Distance-Based Abstraction”, Computer Graphics Forum, vol. 36, no. 3, pp. 227–238, 2017.