Research And Analysis Of Visualization For Software Networks Topology And Parameters

The problems of relevant development and maintenance of large-scale software system brought by its structural complexity has become a key issue that needs to be solved in software quality assurance. The utilization of complex network theory is an effective quantitative means in researching and analyzing static structure complexity of software. Research on topology and network information visualization have been an important method in the research of complex networks. Visualization analysis of software network can get the metrics information of relevant software structure more intuitively and quickly.There are problems in the popularly analysis tool of complex network Gephi that it makes no calculation of some software network metrics parameters(such as ripple degree) and the colored operation is complex that users need to set coloring scheme alone every time. we use C language to achieve efficiency GVA (grid algorithm based on variable) layout algorithm, which can be applied to large-scale software network layout. Then the universal parameter coloring algorithm is used for node coloring according to different parameter values. This coloring algorithm has strong universality, simplifies the cumbersome steps of design about coloring scheme, and provides better and more convenient view assistance for analyzing software system.Because the existing complex network community visualization algorithm has deficiencies of high time complexity, blurred view community boundaries, no clear relationship between community, a new community visualization algorithm is in great need to analyze the complex relationship among internal modules of software network. The experiments show that the algorithm has a good effect of view display, and has lower time complexity.The view colored by metrics parameters shows the distribution of metric parameters, highlighting those nodes of large parameter values. By focusing on these nodes, the developers can measure the complexity of internal structure about software according to different color view. Visualization figure of community structure can clearly show a complex relationship between the communities and prominent effect of modular, avoid the phenomenon that accumulation of nodes in the view of central caused by current community visualization layout algorithm commonly used. Visualization of community structure provides a new perspective for understanding the complex relationship between the internal module of software and has practical significance and application prospects.