New node centrality measurement for directed networks called the Sales Potential is introduced with the property that nodes with high Sales Potential have small in-degree and high second-shell in-degree. Such nodes are of great importance in online marketing strategies for sales agents and IT security in social networks. We propose an optimization problem that aims at finding a finite set of nodes, so that their collective Sales Potential is maximized. This problem can be efficiently solved with a Quasi-parallel Genetic Algorithm defined on a given topology of sub-populations. We find that the algorithm with a small number of sub-populations gives results with higher quality than one with a large number of sub-populations, though at the price of slower convergence. © 2012 Springer-Verlag.
CITATION STYLE
Wang, C. G., & Szeto, K. Y. (2012). Sales potential optimization on directed social networks: A quasi-parallel genetic algorithm approach. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7248 LNCS, pp. 114–123). https://doi.org/10.1007/978-3-642-29178-4_12
Mendeley helps you to discover research relevant for your work.