Research On Propagation Source Location Method In Dynamic Networks Based On Label Propagation

In recent years,complex networks have always been active in various fields of scientific research.With the gradual deepening of research,many researchers begin to pay attention to the problem of source location in complex network.In the real world,there are many phenomena that can be modeled as the spread of propagation sources on complex networks,such as the spread of rumors on social networks represented by microblogs and the spread of respiratory infectious diseases such as tuberculosis on human contact networks,etc.If we can locate the propagation source quickly and accurately,it will be helpful to prevent the further spread of rumor and disease.Therefore,this is a practical work.At present,the research on the location propagation source can be divided into static network and dynamic network based on the tense of the network.Among them,most of the research focuses on the location propagation sources on static networks,only a few focus on dynamic networks,and the precondition is to know the specific propagation model,which is inconsistent with the reality.Based on the above problems,this paper mainly conducts research on the method of locating the propagation source in dynamic networks:Firstly,the method of locating the propagation source based on label propagation in dynamic networks is studied.This work focuses on the problem of locating the propagation source in the case of unknown propagation models on dynamic networks,and locating the source of propagation by hierarchical processing of dynamic networks and label propagation backwards.In the experiment,three different propagation models are used to simulate on four networks.The experimental results prove the effectiveness of the algorithm to locate the propagation source on dynamic networks,and the optimal value range of the experimental parameter is obtained,it is proved that the topology of the network and the information of neighbor nodes have the same influence on the state of nodes.Secondly,the method of locating the propagation source in dynamic networks with node centrality is studied.In order to further improve the accuracy of locating the propagation source on dynamic networks,the centrality algorithm is introduced to screen out important nodes in the infection graph before locating the propagation source.In the experiment,four different centrality algorithms are tested.The experimental results show that the four centrality algorithms can improve the accuracy of locating the propagation source to a certain extent.In addition,it can be seen that the improvement effect of the four centrality algorithms for the same data set is almost the same,which also reflects that the four centrality algorithms are applicable to the problem of source location.This study solves the problem of unknown specific propagation model in real networks,which is a breakthrough for the research of locating propagation source on dynamic networks,making the research of source location more in line with the reality.