An approximate marginal spread computation approach for the budgeted influence maximization with delay

Given a social network of users with selection cost and fixed budget, the problem Budgeted Influence Maximization finds subset nodes ( known as seed nodes) for initial activation to maximize influence, such that total is within allocated budget. Existing solution methodologies this make two assumptions, which are not applicable real-life situations. First, an influenced node current time stamp can trigger only once in next its inactive neighbors other one diffusion process continues forever. To more practical, paper, we introduce Delay by relaxing single triggering constraint imposing additional maximum allowable time. For purpose, consider delay distribution each edge network, influenced, if it so, We first propose incremental greedy strategy solving problem, works based on approximate computation marginal gain influence spread. Next, subsequent improvements algorithm terms efficiency exploiting sub-modularity property delayed function. implement proposed three benchmark datasets. Reported results show set selected lead number compared obtained baseline methods. also observe between improvised methodologies, second efficient larger