Hasso-Plattner-Institut
  
Hasso-Plattner-Institut
Prof. Dr. Jürgen Döllner
  
 

Visualization Techniques for Big Data

Increasing amounts of large, dynamic, heterogeneous, distributed, and complex data raise the question how applications, systems, and users can benefit from this data. Faced with Big Data, most visualization techniques cannot maintain interactive frame rates, produce visual clutter, and are bound to limited memory. Therefore, in this master's thesis aggregation strategies and level-of-detail concepts are developed, evaluated, and integrated, enabling those techniques to handle massive amounts of data. For this, a high-performance hierarchy visualization and rendering framework is provided.

If you are interested or have any further questions please contact the Computer Graphics Systems Group:

Visualization of System Evolution

A treemap is a common visualization technique for hierarchical data with the possibility to map several additional visual attributes (e.g., base area, height, color, face texture). Often the hierarchical data is derived from systems that change over time (e.g., software systems). These changes include both hierarchical structure and related attribute value changes. We provide two master's theses in the domain of hierarchy change visualization. As a starting point we provide a high-performance hierarchy visualization and rendering framework as well as a research prototype.

If you are interested or have any further questions please contact the Computer Graphics Systems Group:

Perception and Comprehension of 2.5D Treemaps

2.5D treemaps are modern visual analysis tools for massive attributed hierarchical datasets. The main goal of this master's thesis is to gain more insights on users' perception and comprehension of single visualization techniques as well as 2.5D treemaps in general. Treatment of this thesis requires a strong focus on rendering, perception, evaluation, and statistics. We offer the possibility to use state-of-the-art a perceptional evaluation environment and hardware (e.g., EyeFollower eye tracking system, virtual reality with Oculus Rift, ...) and  a high-performance visualization and rendering framework.

If you are interested or have any further questions please contact the Computer Graphics Systems Group:

Treemap Layout Stability

The use of treemap layout algorithms depicting varying hierarchies (e.g., over time) are faced by a challenge named "layout stability". That is, small changes in the underlying data should result in small changes of the final layout to preserve a previously created user's mental map. We provide two master's theses focussing on certain aspects of treemap layout stability. Implementations will be based on a framework that allows for the depiction of varying attributed hierarchical datasets with different common 2.5D treemap algorithms.

If you are interested or have any further questions please contact the Computer Graphics Systems Group:

2.5D Treemap Interaction Techniques

Visual analytics and the underlying visualization process require for interactive exploration and navigation techniques. This master's thesis focuses on concepts that increase effective exploration and communication of 2.5D treemaps. A variety of interesting application scenarios exist, that can be elaborated: For example, large multi-touch screens used for collaborative exploration (collaboration wall); focus+context concepts for filtering of massive information; or virtual reality environments for distraction-free data examination.

If you are interested or have any further questions please contact the Computer Graphics Systems Group: