Design, Development and Evaluation of a Big Data Analytics Dashboard

This Master’s thesis focuses on the Design, Development and Evaluation of a novel Visual Analytics Dashboard for Big Data Analytics. The presented dashboard connects social activity from Facebook with a thorough event timeline of the factory disasters in the Bangladesh garment industry. Bangladesh depicts one of the largest garment industries in the world, and their mostly female workers only receive a low wage. The goal of this thesis is to present a thorough understanding of the design and development processes needed to implement a Big Data Visual Analytics tool based on freely available open-source components in a robust, extensible manner. Moreover, an evaluation of the developed dashboard is performed based on a task-based user study in conjunction with software and database performance optimization. The user study concludes that the dashboard is easy to use in a productive manner without prior training and experience in using visual analytics tools. By using the presented dashboard, even novice users can gain profound understanding of the tragedies in Bangladesh, their background, and the resulting social media impact. Furthermore, the linked social media activity from eleven international companies in the garment industry can be interactively explored through different visualizations depicting actor mobility, conversation content, language distribution, and overall activity levels.