Maximizing the Influence: Analytic Activation Probability Computation Approach

TL;DR

This paper introduces a novel analytic activation probability computation method for influence maximization, improving accuracy and efficiency, enabling scalable solutions for large social networks.

Contribution

It presents a new nonlinear system for more accurate activation probability estimation and an efficient computation method, advancing influence maximization algorithms.

Findings

Outperforms previous methods in accuracy of activation probabilities

Reduces running time significantly, enabling scalability

Demonstrates effectiveness on real-world datasets

Abstract

Influence maximization is the problem of finding a subset of the most influential individuals in a network. The impact of social networks on the dissemination of information and the development of viral marketing has made this problem as the subject of many studies. Influence maximization is $\NP$-hard and many greedy algorithms have been proposed to solve the problem. In this paper we propose a greedy algorithm that approximates the influence by using a novel analytic activation probability computation method. We propose a nonlinear system of equations to compute the activation probabilities which is more accurate than state-of-the-arts. Moreover, we propose a method to compute the activation probabilities efficiently, in order to reduce the running time of the algorithm. We examine our proposed methods on some real-world data sets. Our experiments demonstrate that the proposed algorithms outperform the previous works by computing more accurate activation probabilities. In addition, Our efficient method has much improvements on the running-time, so it can be easily scaled on large networks.