A holistic exploration of influence maximization, opinion maximization, and opinion dynamics in social networks: Methodological survey and challenges

Document Type : -

Authors

1 Imam Hossein comperhensive University,Tehran,Iran

2 Imam Hossein comprehensive University, Tehran, Iran

3 Faculty of Imam Hossein comprehensive University, Tehran, Iran

Abstract

Analyzing data from online social networks has emerged as a significant scientific challenge. Among the key research areas within this domain are influence maximization, opinion maximization, and dynamic opinion maximization. Influence maximization aims to identify an initial set of users (seed nodes) to maximize the spread of information or messages within a social network. Opinion maximization focuses on increasing the number of users holding positive opinions, thereby enhancing the overall opinions across the network. Dynamic opinion maximization, which accounts for the time-varying nature of user opinions, seeks to maximize the proportion of users with positive opinions over time. This paper survey the existing research on influence maximization, opinion maximization, and dynamic opinion maximization, critically examining the methods proposed in previous studies. It compares these methods in terms of their strengths, weaknesses, and challenges. Furthermore, a comparative study is conducted using available datasets related to opinion maximization to evaluate the performance of these methods. The primary contribution of this research is the creation of a comprehensive comparison table, outlining the strengths and weaknesses of existing methods. The results emphasize that maximizing positive opinions remains a central objective in this field, and highlight the challenges posed by the dynamic nature of opinions in social networks.

Keywords