The Research And Implementation Of Rapid Information Propagation Algorithm In Social Network

In recent years,the research on social network for information propagation has made considerable progress.These research directions include determining the structure of social network,researching information propagation model,researching how to suppress or accelerate information propagation and positioning rumors in social network.As social network widely used in people's daily life,to obtain information relying on social network has become an important means of information acquiring.Thus,it is significant to find effective methods to speed up the propagation of information.There are three standard for measuring the fast information propagation.The first is the propagation time,the shorter the time for all nodes in the network receiving information is,the faster the speed of propagation will be.The second is propagation range,there are a lot of information which cannot be propagated throughout the network.So if the coverage of information is wider in the same period of time,the speed of information propagation will be faster.The final factor is the cost of propagation.The more the amount of the initial activation node selected is,the higher the cost of spread will be.In this paper,we propose relevant fast information propagation algorithm from two aspects after analyzing the advantages and disadvantages of traditional fast information propagation algorithm.The first aspect is the set of initial activation nodes selected according to traditional fast information propagation algorithm has the problem that a large number of nodes are too concentrated in the network.For example,when two most important nodes in social network have connected side,we should not select these two nodes as initial activation nodes.To solve this problem,we propose a hierarchical network node selection method and dispersedly select nodes with high centrality according to network hierarchy thinking.The second aspect is how to evaluate the importance of remaining network nodes after a node was selected.For instance,when a network node was selected,excluding this node,the topology of remaining nodes will change.At the same time,the evaluation of the importance of remaining network nodes also will change.We propose node selection method of restructuring network to evaluate the importance of the remaining nodes as accurate as possible.Finally this paper also implements fast information propagation method and designs this method to be a desktop tool.With this tool,we can select initial activation nodes in a better way and implement faster information propagation in social network.A large amount of experimental results show that compared with traditional method of network node selection,node selection method based on hierarchical network and selection method based on network restructuring can make initial activation nodes selection effective.Which can be conducive to the fast propagation of information in social network.