Research On Layout Algorithm For Complex Networks

Complex network visualization aims to transform complex networks into intuitive graphics,and add corresponding visual interaction technology to help users understand,analyze and explore complex networks,and has a wide range of applications in various disciplines and fields.Among them,the network layout algorithm is the basis of visualization,and its layout results directly affect the quality of the visualization,and also affect the user's understanding of the network structure.The increase in the scale of network data brings challenges to both the layout efficiency and the layout quality.In the face of large-scale networks,the layout of existing algorithms is difficult to take into account the overall structure and local details of the network.Therefore,starting from the community structure of complex networks and based on the idea of multi-level layout,this paper proposes a hierarchical layout algorithm for complex network clustering features to take into account the overall and local structure of the network.Furthermore,in order to meet the needs of dynamic network layout,this method is extended to dynamic networks,and a new node movement restriction scheme is designed.The specific work is as follows:(1)Network clustering pattern discovery is an important task in network analysis.A good network layout as a whole should reflect the clustering characteristics in the network and allow users to explore the network structure from different levels.Therefore,based on the idea of community detection and multi-level layout,a hierarchical layout algorithm of clustering features is proposed.Firstly,the network is divided by seed nodes and personalized Page Rank;secondly,the network is coarsened according to the detection results,and the initial layout of the coarsened network is designed;then the force-directed model is improved by using the node degree information to complete the refinement.Finally,in order to verify the effectiveness of the algorithm,experiments from the whole to the local are designed.The results show that the algorithm can generate high-quality layout results in an effective time.Compared with the existing layout algorithms,the algorithm can more realistically display the network clustering characteristics as a whole,and at the same time better reflect the local structural shape and characteristics.(2)Different from the static network layout,the dynamic network layout requires maintaining the user's mood map while ensuring certain aesthetic standards,so that the user can discover and understand the structural changes of the network when the network is updated.In order to meet the needs of dynamic network layout,based on the static network layout algorithm,steps such as initial network layout and location determination of new nodes are added to extend it to dynamic networks.In order to maintain the user's mood map and make the layout result have high quality,a new node movement restriction strategy is designed.The strategy considers the degree of influence of network structure changes on different nodes,and also considers the age of nodes.In this way,a balance is struck between maintaining the mood map and the quality of the layout.The experimental results show that the layout results of the algorithm have high layout quality and maintain the stability of the layout to a certain extent.(3)In order to further help users understand and analyze network data,an interactive complex network visual analysis prototype system is designed and implemented based on the above algorithm.In order to better meet the analysis needs in different scenarios,the system integrates a variety of commonly used network layout algorithms.Combined with a variety of visual interaction technologies and multi-view collaborative analysis technologies,it gives users the ability to manipulate data,which is convenient for users to analyze and explore data.The system evaluation shows that the system has high availability and effectiveness.