Selecting influential nodes for detected communities in real-world social networks

The problem of influence maximization is to find initial users in social networks so that they eventually influence the largest number of people. This problem is used in wide areas such as epidemiology, economics for detecting the spread of an infection disease, marketing a new product as quickly as possible, respectively. We propose three heuristic algorithms for influential nodes selection after detecting communities in social networks. They are faster than an original greedy algorithm and close to its influence spreads. We evaluate influential nodes selection algorithms on a large academic collaboration network. We experimentally demonstrate that our proposed algorithms outperform the greedy algorithm and traditional heuristic.