A related topic is that of random molecular networks, which serve as benchmarks against which data mining results are measured.
For the social network control based on random topics, we assume that, with a given frequency, a certain subset of agents is selected and a variable number of random topics, among the ones existing in the network, are inserted into their Personal state with an average level of interest.
Label C0 represents the results of the natural system dynamics with no addition of random topics.
Results of simulations are presented in Figs. 1 and 2, which show, respectively, the network and the diffusion effects of the strategic use of random topics.
This signals that more triangles have formed, a direct effect of the new connections produced by the introduction of random topics in hub nodes.
The injection of random topics, instead, in all situations, even for very few driver nodes (e.g., for (N=100), the setting HIGH 1% means that the node with the highest degree is the only driver agent) is able to produce a sensible reduction of the average degree with respect to the benchmark C0.
For the network effect, Fig. 1 presents how the average node degree and the clustering coefficient change with respect to the benchmark represented by the typical network behavior with no injection of random topics in nodes (label C0 in Fig. 1) for three network sizes ((N=[100,500,1000])).
The key parameters for defining a social control strategy are selection criteria of driver nodes, number of random new topics introduced, and frequency of introduction of new topics.
References regarding other related topics, including image analysis, modeling of random objects, and space-time analysis, are provided.
From the point of view of classical Banach space theory, it is a quite natural topic to give a reasonable definition of random smoothness in random normed modules.
