Pseudo-Cores: The Terminus of an Intelligent Viral Meme’s Trajectory

Abstract

Comprehending the virality of a meme can help scientists address problems pertaining to disciplines like epidemiology and digital marketing. Therefore, it does not come as a surprise that meme virality stands out as an integral component of research in complex networks, today. In this paper, we explore the possibility of artificially inducing virality in a meme by intelligently directing a meme's trajectory in the network. Keeping in mind the importance of core nodes in a core-periphery structure, we propose two shell-based hill climbing algorithms to determine the path from a node in the periphery shell (where the memes generally originate) to the core of the network. On performing further simulations and analysis on the networks behavioral characteristics, we were also able to unearth specialized shells which we termed Pseudo-Cores. These shells emulate the behavior of the core in terms of the size of the cascade triggered. In our experiments, we have considered two sets for the target nodes, one being core and the other being any of the pseudo-cores. We compare our algorithms against already existing path finding algorithms and validate the better performance of our algorithms experimentally.

Evolving models for meso-scale structures

“…Table 1 shows that the core constitutes less than 2 percent of the whole network. Gupta et al [47] observed that few innermost shells close to nucleus have the same influential power as core and have very high density. In practical use, a very small percentage of nodes from innermost layers can also be considered as core nodes.…”

Maximising Information Diffusion with Source and Locality Awareness

“…al. has referred such nodes as "Pseudocores" [46] in their work. Our experiments demonstrate the disappearance of pseudo-cores when the edge weights are considered in the diffusion process.…”

Core-based Approach to Measure Pairwise Layer Similarity in Multiplex Network

“…In Layer 1 (L 1 ) and Layer 2 (L 2 ), the list of core nodes is [9,2,7,3,5,4,11,15,16,13] and [3,5,6,9,11,4,7,8] respectively. The common set of nodes S between L 1 and L 2 is [3,4,5,7,9,11], and union between the two lists is [2,3,4,5,6,7,8,9,11,13,15,16]. The Coreness SimðL 1 ; L 2 Þ ¼ 0:5 by using Eq.…”

“…As the study on monoplex network [4] reported existence of core-periphery structure [6] in real world network, the core of such networks plays an important role in maximizing the diffusion in the network [7]. In multiplex network, each layer has its own core-periphery structure.…”