Abstract:Influence maximization is the task of finding k seed nodes in a social network such that the expected number of activated nodes in the network (under certain influence propagation model), referred to as the influence spread, is maximized. Lattice influence maximization (LIM) generalizes influence maximization such that, instead of selecting k seed nodes, one selects a vector x = (x_1, ..., x_d) from a discrete space X called a lattice, where x_j corresponds to the j-th marketing strategy and x represents a marketing strategy mix. Each strategy mix x has probability h_u(x) to activate a node u as a <a class="link-external link-http" href="http://seed.LIM" rel="external noopener nofollow">this http URL</a> is the task of finding a strategy mix under the constraint x_1+...+x_d <= k such that its influence spread is maximized. We adapt the reverse influence sampling (RIS) approach and design scalable algorithms for LIM. We first design the IMM-PRR algorithm based on partial reverse-reachable sets as a general solution for LIM, and improve IMM-PRR for a large family of models where each strategy independently activates seed nodes. We then propose an alternative algorithm IMM-VSN based on virtual strategy nodes, for the family of models with independent strategy activations. We prove that both IMM-PRR and IMM-VSN guarantees 1-e-\epsilon approximation for small \epsilon> 0. Empirically, through extensive tests we demonstrate that IMM-VSN runs faster than IMM-PRR and much faster than other baseline algorithms while providing the same level of influence spread. We conclude that IMM-VSN is the best one for models with independent strategy activations, while IMM-PRR works for general modes without this assumption. Finally, we extend LIM to the partitioned budget case where strategies are partitioned into groups, each of which has a separate budget, and show that a minor variation of our algorithms would achieve 1/2 -\epsilon approximation ratio with the same time complexity.

Bring Order into the Samples: A Novel Scalable Method for Influence Maximization

Bring Order into the Samples: A Novel Scalable Method for Influence Maximization (extended Abstract).

Improving Influence Maximization from Samples: an Empirical Analysis.

QuickIM: Efficient, Accurate and Robust Influence Maximization Algorithm on Billion-Scale Networks.

Scalable Influence Maximization with General Marketing Strategies.

Scalable Lattice Influence Maximization

CoFIM: A Community-Based Framework for Influence Maximization on Large-Scale Networks

Influence Maximization in Hypergraphs by Stratified Sampling for Efficient Generation of Reverse Reachable Sets

Real-time Targeted Influence Maximization for Online Advertisements

Maximizing Multifaceted Network Influence.

Multi-Round Influence Maximization

Network Inference and Influence Maximization from Samples

BaCIM: Balanced Competitive Influence Maximization Based on Blocked Reverse Influence Sampling.

Multi-Round Influence Maximization (extended Version).

gIM: GPU Accelerated RIS-based Influence Maximization Algorithm

Influence Maximization Revisited: Efficient Reverse Reachable Set Generation with Bound Tightened

LAIM: A Linear Time Iterative Approach for Efficient Influence Maximization in Large-Scale Networks

Fast Graph Centrality Computation Via Sampling: a Case Study of Influence Maximisation over OSNs.

Influence maximization based on community structure and second-hop neighborhoods

A Linear Time Algorithm For Influence Maximization In Large-Scale Social Networks

Boosting Information Spread: An Algorithmic Approach