Exploration and Analysis of Flow Data in Augmented Reality

Kurzfassung

Generating information from data is a constant challenge in information technology. The advent of modern data collection techniques has led to a significant increase in data generation, resulting in the accumulation of huge amounts of data. Flow data can be generated in different ways. Simulations are used in a variety of research areas to generate datasets. This paper explores the potential for visualising large datasets in augmented reality. The Microsoft HoloLens 2 enables the interactive exploration of data. This thesis examines the pre-processing of data sets, their required structural design, the resulting organisation and various possibilities for visualising this data. The development of acceleration structures enables independent exploration of data sets without connection to other devices. At the centre of the exploration is the concept of individually configurable focal points, which can vary depending on the subject area. Specific features of the data flow are identified and visualised. In order to utilise the developed acceleration structures efficiently, an interaction method is selected and implemented. This enables a fluid visualisation depending on the size of the data sets. The presentation of the data is characterised by the focus and context principle, which provides information about the environment in addition to the most important data. The decision on relevance is made on the basis of multidimensional factors that can be used to determine the ranking of the data. The various data structures and the ranking solution were tested for their efficiency. The trade-off between performance and data loss is analysed in detail and underpinned with figures. Practical use cases were used to demonstrate that searched structures can be found in the data records using the application.