Analysis of Various Graph Layout Approaches Used in GUESS Software

Himanshu Sharma, Vishal Srivastava

One challenge in node-link diagrams is how to efficiently provide a node placement or layout that will yield a meaningful graph visualization. For simple structures, the system needs only a set of aesthetic choices to provide a useful graph—sometimes even a hand-drawn visualization could suffice. But for large, complex structures, effective layouts are harder to create, which motivates continual interest in graph layout algorithms as an integral part of visualizing complex networks. Although most traditional work involves developing more efficient layout methods for static graphs, more recent efforts have also focused on finding effective ways to generate dynamic graphs of time-varying networks. This paper discusses the various aesthetic criteria's which improve the readability of graphs and helps in how to choose proper layout algorithm for specific data to make the visualization better. This paper also discusses the different graph layout approaches which are used as the basis for developing many other new and improved graph layout algorithms helping in better visualization of graphs. Some easy-to-program network layout approaches are discussed here, with details given for implementing each one. This paper is mainly focused on the basic graph layout approaches which are used in "GUESS"  the graph visualization and exploration software. This paper is also intended to beginners who are interested in programming their own network visualizations, or for those curious about some of the basic mechanics of graph visualization.