Abstract not found

This paper presents the framework of a scale-oriented scheme for the presentation and classification of reverse engineered sections of procedural code into objects. The aim is to develop an extensible system framework, which allows the output from a suite of data analysis tools to be visually presented to a user. The relationship between the analysis and visualisation is a progressive cycle, where each time through the cycle the overall quality of the classified objects improves. This framework supports two distinct methods of information feedback from the visualisation to the analysis suite. The two feedback loops aim to increase both the ease of understanding for the reverse engineer and the quality of the resultant objects. As the analyst views the visualisation the perceived view of the relationships exhibited in the system may be modified, removed or added to. This results in a change to the underlying graph or the clustering of that graph, which must be addressed in the visual presentation of the information using a variety of techniques to maintain the users `mental map', or understanding each time through the cycle.

Code conversion tools in Reverse Engineering perpetuate the legacy code problem; the new code does not contain any new understanding of what is occurring in the system. In attempting to aid system comprehension, current research focuses on extracting a model or representation of the system from the code. From this model, reverse engineers try to form a picture of the original design. Our approach hopes to aid the reverse engineer, in moving from model to design by presenting a concise and useful graphical representation of the model.

Typically, a user can understand a clear textual description of a small amount of relational information. However, when dealing with large amounts of relational information, a vast amount of text quickly becomes cumbersome. Existing methods and techniques, for the visualization of relational information, tend to effectively deal with only relatively small data sets. Here we present methods to generate a hierarchical compound graph for the large 3D visualisation of relational information. These large compound graphs can effectively be drawn using a modified force-directed algorithm with horizon drawings.

This paper describes the FADE paradigm for visualization and a series of experiments for the fast layout, abstract representation, and measurement of software views. In program comprehension, graph models are typically used to represent relational information, where the visualization of such graphs is referred to as graph drawing. Here we present the results of an investigation into efficient techniques for drawing and abstractly representing large software views with thousands of nodes from four medium sized software systems. The paradigm presented in this paper marries a solution to problems of computation time, screen space, cognitive load, and rendering for large-scale drawings using a single graph model.

Graph visualization has been widely used to understand and present both global structural and local adjacency information in relational datasets (e.g., transportation networks, citation networks, or social networks). Graphs with dense edges, however, are difficult to visualize because fast layout and good clarity are not always easily achieved. When the number of edges is large, edge bundling can be used to improve the clarity, but in many cases, the edges could be still too cluttered to permit correct interpretation of the relations between nodes. In this paper, we present an ambiguity-free edge-bundling method especially for improving local detailed view of a complex graph. Our method makes more efficient use of display space and supports detail-on-demand viewing through an interactive interface. We demonstrate the effectiveness of our method with a public coauthorship network data.