Research Of Temporal Network Visualization Method Based On Adaptive Time Slices Partition

Isometric time slice is commonly used to visualize the evolution of the network for the present temporal network visualization methods.However,isometric time slice has many disadvantages,such as the difficulty of obtaining time granularity,the poor data truth reflected by segmentation result and the inconvenience of study and discovery for temporal patterns.In view of the above problems,this paper studies the temporal network visualization method based on adaptive time slices partitioning.The main research contents are as follows:(1)To improve the efficiency of the temporal network for users meanwhile to effectively display the structural characteristics and changing patterns of the network,this paper proposes an Adaptive Time Slice Partition method based on the inherent characteristics of the temporal network.The method can cluster temporal network data in time dimension on the condition of the inherent properties.On the basis of two representations of temporal networks(event-based representation and snapshot representation),An adaptive time slice representation model is designed and an improved model to describe events whose interval obeys the long-tail distribution is established.At the same time,time slices are defined for different exploration tasks,this paper also proposes adaptive time slice partition rules based on temporal patterns and center nodes.What is more,implementation algorithms including Snapshot Label algorithm,Hierarchical Partition Based on Temporal Pattern algorithm and Incremental Partition Based on Temporal Center Node algorithm are put forward.(2)To effectively display the unequal time slices obtained by Adaptive Time Slice Partition method,this paper proposes a different visualization framework by using the animation and juxtaposition visualization methods,and juxtaposing the unequal time slices of the temporal network It is convenient for users to quickly capture the changing patterns of the temporal network.At the same time,it provides users with more detailed time slice development by combining animation or flipbook.For this reason Simulated Annealing Layout Based on Aesthetic Standards algorithm and Simulated Annealing Layout Based on Mental Map are put forward.(3)The design and implementation of temporal network visualization system based on adaptive time slice partitioning is carried out.In order to verity the effectiveness of temporal network visualization method based on adaptive time slice partitioning,the paper validates and analyzes the efficiency and effect of time slice partitioning and layout algorithm,and temporal network based on adaptive time slice partitioning.Detailed experimental analysis was carried out on the practical application of the visualization method.The results show that the Adaptive Time Slice Partition method can represent temporal characteristics of the network more accurately than the traditional time slice partitioning method.The simulated annealing layout algorithm based on aesthetic criteria and mental map can increase the algorithm efficiency and improve layout quality.The network characteristics can be effectively summarized and displayed by temporal network visualization method based on adaptive time slice partitioning,and the visual analysis efficiency can be improved.