嵌入社区半径的力引导与径向树混合布局算法

Community radius embedded in force-directed and radial tree hybrid layout algorithm

力引导布局算法存在无法展示复杂网络社区结构的缺陷,虽引入聚类的方式来展示社区结构,但社区内节点拥挤且排列无序,不利于观察社区内节点的结构特征与连边关系,为此提出嵌入社区半径的力引导与径向树混合布局算法.该算法首先采用K-means算法对网络节点进行社区划分;然后,用社区内节点数量确定社区半径,并将社区半径嵌入到社区斥力、引力中来展示社区结构;最后,采用径向树布局分层可视化各社区内节点.实验中使用拥挤区域占比、点分布偏差、节点偏差等指标验证了本算法既能降低拥挤度又能减少节点布局偏差,可视化结果显示,本算法布局社区结构明显,节点层次分明,易于理解.

Force-directed layout has a defect that can't display complex network community structure.Although cluster layout algorithm is introduced to visualize community structure,nodes in the community are crowded and disordered,which is not conducive to observing structural features and nodes connection relationship in the community.Therefore,community radius embedded in force-directed and radial tree hybrid layout algorithm is proposed.Firstly,the algorithm uses the K-means algorithm to divide the network nodes into community.Then,the community radius is determined by the number of nodes in each community,and the community radius is embedded into the repulsion and gravity to achieve the effect of cluster layout.Finally,radial tree layout is used for each community to hierarchically visualize the nodes within the community.In the experiment,crowed area ratio,point distribution deviation,node deviation and other indicators are used to show that the algorithm can reduce congestion and node layout deviation.Visual results demonstrate that layout structure of the algorithm is obvious,and nodes are clearly structured and easy to understand.