Information Influence Maximization Algorithm Based On Cost Budget

Social network is an important product in the information age,and the data in the network grows exponentially every day,how to effectively analyze and process this data,is researched in social networks,and influence maximization is one of the important research directions.Influence maximization involves all aspects of daily life,for example:product promotion,information security,web search,etc,in real life,the influence maximization is usually related to cost budgeting.Therefore,maximization under the influence of cost budget has important research significance.in this paper,following research will be carried out by studying the influence maximization algorithms and information propagation models,and combining the relevant theories of graph theory.First,based on the problem of node cost in the network,an algorithm for node influence maximization based on cost budget is proposed,The first algorithm proposed in this paper first analyzes the node’s KS index and structure hole attributes,As well as the betweeness centrality and clustering coefficient of nodes,through the balance coefficient,the internal and external attributes of the node are organically combined,so that these attributes better reflect the importance of the node,in addition,combining the node cost and node attributes,a new propagation probability is proposed,which maximizes the influence of the node and at the same time,the propagation of information is more in line with the real situation.Secondly,based on the propagation cost problem in the network,a road optimization algorithm based on cost budget is proposed,this algorithm is based on the idea of Dijkstra’s algorithm,which achieves the minimum path cost while propagating information to the final node path,the algorithm first analyzes the calculation method of edge weights,and based on the propagation cost and response time of edges in weighted graphs,a new method for calculating edge weights is proposed,as well as combining the importance of nodes,the concept of equivalent path is proposed.The algorithm can achieve optimal path under the premise of cost budget.Finally,comparative experiments on the two proposed algorithms,use different types of network datasets in the experiment,and compare with the same type of algorithm,through comprehensive analysis of experiments,it can be seen that although the running time of the two algorithms is not the shortest,it has the largest influence range and the overall effect is ideal.