Neuronale Netze im Virtuellen Raum - Entwicklung einer benutzerfreundlichen Virtual Reality Anwendung zur Exploration und Datenkommunikation Künstlicher Neuronaler Netze

Kurzfassung

Machine Learning (ML) and Artificial Neural Networks (ANN) are used in many areas of industry and science today. In the course of digitalization, these technologies will become even more relevant and open up new areas for themselves. This will probably also increase the need for effective learning methods and alternative forms of representation to achieve a higher degree of understanding of algorithms and processes within KNN. A complementary form of representation or learning environment could be for example Virtual Reality (VR), as the technology offers new possibilities and perspectives regarding representation and interaction. This thesis investigates to what extent the use of a VR application for exploration of KNN could contribute to the knowledge gain of KNN. Furthermore, it is investigated which requirements there are for such a VR application with regard to content components and user-friendliness. In the context of this work, requirements were determined with the help of a literature research and a survey. In consideration of the requirements a concept was developed, which served as framework for the implementation of the VR application. The concept was implemented and evaluated as a prototype using an appropriate development methodology. The results of the evaluation show that the VR application can be operated in a user-friendly way and could probably lead to a gain in knowledge. Further studies in this area should follow in order to validate the assumption of the knowledge gain on the one hand and to examine the VR application more closely with regard to its usability on the other hand.