Optimizing Spread of Influence in Weighted Social Networks via Partial Incentives

TL;DR

This paper explores influence spread in weighted social networks, introducing cost and incentive considerations, providing algorithms and hardness results, and validating findings through extensive simulations on real networks.

Contribution

It extends influence diffusion models by incorporating node costs and incentives, offering new algorithms and complexity results, and validating them experimentally.

Findings

Incentives reduce total influence costs in networks.

Algorithms effectively identify minimal-cost influence sets.

Experimental results confirm theoretical predictions on real networks.

Abstract

A widely studied process of influence diffusion in social networks posits that the dynamics of influence diffusion evolves as follows: Given a graph $G=(V,E)$, representing the network, initially \emph{only} the members of a given $S\subseteq V$ are influenced; subsequently, at each round, the set of influenced nodes is augmented by all the nodes in the network that have a sufficiently large number of already influenced neighbors. The general problem is to find a small initial set of nodes that influences the whole network. In this paper we extend the previously described basic model in the following ways: firstly, we assume that there are non negative values $c(v)$ associated to each node $v\in V$, measuring how much it costs to initially influence node $v$, and the algorithmic problem is to find a set of nodes of \emph{minimum total cost} that influences the whole network; successively, we study the consequences of giving \emph{incentives} to member of the networks, and we quantify how this affects (i.e., reduces) the total costs of starting process that influences the whole network. For the two above problems we provide both hardness and algorithmic results. We also experimentally validate our algorithms via extensive simulations on real life networks.