Model-based development of visual data analysis applications

OData support
Gönczy László
Department of Measurement and Information Systems

The goal of my thesis is to give a method for generating R based, interactive exploratory data analysis and visualization applications from a declarative, XML-like language using multiple technologies. R is a programming language used for statistical calculations and data analysis, but thanks to its extensions it is also excellent for data visualization. In order to extract the important information from the visualized data easier and faster, it is useful to develop an application which supports dynamic manipulation and has a graphical user interface. These applications include many typical functions, like handling of widgets, plotting and data table operations, therefore to accelerate development it is possible to create a code generator which covers the whole application. This approach makes it also possible to create multiple analysis and visualization programs implemented in different technologies, from a high level, mostly technology independent set of declarations. In my thesis I first overview the tools of exploratory data analysis, graphical user interfaces, and R based technologies which implement the former. After this I define an XML based language which is suitable to describe these functions simply. I write the software which generates the R realizations automatically. I verify its operation on example programs. The method should be implemented in such a way, that the high level declaration is easily extendable, and the code generator can be simply supplemented with the parts that realize the newly added language features. To demonstrate heterogeneity there is at least one presented example that uses thick client technology and one that builds on web based client-server architecture.


Please sign in to download the files of this thesis.