A reusable framework to accelerate the development of visual analytics applications based on dimensionality reduction

To be able to visualize and analyze multidimensional data sets, the dimensions have to be reduced to two or three, by using dimensionality reducing (DR) methods. The problem is that there is no established foundation for building applications to handle data sets, DR and visualization and therefore the process of doing so is not as efficient as it could be. This thesis suggests a framework as a solution to the problem, and covers the implementation of such a framework. The framework is used to create a prototype application to ensure that it is useful in such a scenario. For further evaluation, a domain expert tested the framework by following the associated instructions, and answered a questionnaire. The answers were positive and contained comments that were used to improve the instructions, and suggestions on how to improve the framework in the future.