Composite Community-Aware Diversified Influence Maximization with Efficient Approximation

TL;DR

This paper introduces a novel influence maximization approach that considers multiple community structures to enhance diversity, employing a new sampling technique and a two-stage algorithm for efficient approximation.

Contribution

It formulates the composite community-aware diversified IM problem, proposes a new sampling method and a two-stage algorithm to efficiently approximate solutions with theoretical guarantees.

Findings

The proposed G-HIST algorithm achieves a $(1-1/e- ext{varepsilon})$ approximation.

G-HIST significantly reduces sampling complexity and improves efficiency.

Experimental results demonstrate the algorithm's effectiveness and superiority over baselines.

Abstract

Influence Maximization (IM) is a famous topic in mobile networks and social computing, which aims at finding a small subset of users to maximize the influence spread through online information cascade. Recently, some careful researchers paid attention to diversity of information dissemination, especially community-aware diversity, and formulated the diversified IM problem. The diversity is ubiquitous in a lot of real-world applications, but they are all based on a given community structure. In social networks, we can form heterogeneous community structures for the same group of users according to different metrics. Therefore, how to quantify the diversity based on multiple community structures is an interesting question. In this paper, we propose the Composite Community-Aware Diversified IM (CC-DIM) problem, which aims at selecting a seed set to maximize the influence spread and the composite diversity over all possible community structures under consideration. To address the NP-hardness of CC-DIM problem, we adopt the technique of reverse influence sampling and design a random Generalized Reverse Reachable (G-RR) set to estimate the objective function. The composition of a random G-RR set is much more complex than the RR set used for the IM problem, which will lead to inefficiency of traditional sampling-based approximation algorithms. Because of this, we further propose a two-stage algorithm, Generalized HIST (G-HIST). It can not only return a $(1-1/e-\varepsilon)$ approximate solution with at least $(1-\delta)$ probability, but also improve the efficiency of sampling and ease the difficulty of searching by significantly reducing the average size of G-RR sets. Finally, we evaluate our G-HIST on real datasets against existing algorithms. The experimental results show the effectiveness of our proposed algorithm and its superiority over other baseline algorithms.