Big Data Visualization in Extended Reality

Abstract: Effectively utilizing Big Data can be a challenge for many companies. Due to its volume, many aspects of the data processing pipeline such as storage, retrieval and usage may quickly become problematic. For example, it could be difficult identifying the relevant metrics when analyzing a service failure. Visualizing and deriving insights from the data could also prove to be quite onerous. Immersive analytics is one field that aims to solve some of the issues around Big Data by using technologies such as virtual reality, augmented reality and mixed reality. For our thesis project, we have worked with Axis Communications, a Swedish multinational company within the video surveillance industry to design an immersive analytics application for their R&D Tools department. To define our scope, we employed a co-design approach together with an iterative Double Diamond design process. As a result of academic and user research, concept development and iterative lo-fi prototyping, we decided to implement a hi-fi prototype of an application that would help engineers physically navigate their data. In an extrema identification task, our prototype allowed users to navigate the data more quickly, but in terms of accuracy, users performed roughly the same across both tools. Given that this was a relatively simple task, more research would be needed to investigate the performance of immersive analytics applications in complex data analysis operations.