Balance Research on Large-scale Software Systems Based on the K-core Decomposition
We abstract the object-oriented software systems to software coupling network (SCN) with complex features. We analytically study SCNs using the k-core decomposition. Based on the statistics, we analyze the SCNs from different points of view, and important properties of the SCNs are found. In each the middle layers, the degree number of keeps balance well, which is of great significance for the software design and judge. Besides, some other conclusions are draw. The software systems with larger scale often have deeper hierarchy, and the degree number of kcore (the final) is also larger, the ratio between clustering coefficient of the k-core graph and that of original graph is larger, while the node number of k-corc (the final) is not increasing with the scale. The distribution of the k-core node number in every decomposition step complies with exponential distribution.
k-core object-oriented complex network balance hierarchy
Zhenhua Bai Yang Gao Xinjian Zhuo Guoai Xu Miao Zhang
School of Science Beijing University of Posts and Telecommunications Beijing, China Key Laboratory o Key Laboratory of network and information attack & defence technology of MOE Beijing University of P School of Science Beijing University of Posts and Telecommunications Beijing, China
国际会议
成都
英文
184-188
2011-01-14（万方平台首次上网日期，不代表论文的发表时间）