Unveiling Influence in Networks: A Novel Centrality Metric and Comparative Analysis through Graph-Based Models

Nada Bendahman,Dounia Lotfi
DOI: https://doi.org/10.3390/e26060486
IF: 2.738
2024-06-01
Entropy
Abstract:Identifying influential actors within social networks is pivotal for optimizing information flow and mitigating the spread of both rumors and viruses. Several methods have emerged to pinpoint these influential entities in networks, represented as graphs. In these graphs, nodes correspond to individuals and edges indicate their connections. This study focuses on centrality measures, prized for their straightforwardness and effectiveness. We divide structural centrality into two categories: local, considering a node's immediate vicinity, and global, accounting for overarching path structures. Some techniques blend both centralities to highlight nodes influential at both micro and macro levels. Our paper presents a novel centrality measure, accentuating node degree and incorporating the network's broader features, especially paths of different lengths. Through Spearman and Pearson correlations tested on seven standard datasets, our method proves its merit against traditional centrality measures. Additionally, we employ the susceptible–infected–recovered (SIR) model, portraying virus spread, to further validate our approach. The ultimate influential node is gauged by its capacity to infect the most nodes during the SIR model's progression. Our results indicate a notable correlative efficacy across various real-world networks relative to other centrality metrics.
physics, multidisciplinary
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to identify influential nodes in social networks. Specifically, the author proposes a new centrality measurement method - Centrality Degree Paths (CDPs), aiming to more comprehensively evaluate the influence of nodes in social networks. Traditional centrality measurement methods, such as degree centrality, closeness centrality, betweenness centrality and PageRank, etc., are effective but have limitations when dealing with complex networks, for example, over - relying on direct connections or ignoring the influence of indirect paths. ### Main problems and goals of the paper 1. **Identifying key nodes in social networks**: - Traditional methods mainly focus on the number of direct connections of nodes (degree centrality) or other local features, but these methods may overlook the influence of nodes on the entire network through indirect paths. - This paper proposes a new centrality measurement method - CDPs, which considers not only the number of direct connections of nodes but also the indirect influence of nodes through paths of different lengths. 2. **Improving the limitations of traditional centrality measurement**: - Degree centrality is only based on the number of direct connections of nodes, ignoring the influence of indirect paths. - The computational complexity of closeness centrality is high in large - scale networks and is not practical. - Betweenness centrality mainly focuses on the shortest path and may miss other important influence paths. - PageRank is suitable for citation or web page networks, but it is not effective in social networks because the interaction patterns in social networks are more complex. 3. **Verifying the effectiveness of the new method**: - The author verifies the effectiveness of the CDPs method on seven standard data sets through Spearman and Pearson correlation tests. - Further use the SIR model (Susceptible - Infected - Recovered model) to simulate virus spread to evaluate the influence of nodes. The results of the SIR model show that the CDPs method performs well in identifying key nodes. ### Characteristics of the new method - **Comprehensively considering direct and indirect paths**: The CDPs method enhances the influence of highly - connected nodes by squaring the degree of nodes and considers the influence of indirect paths by calculating the number of paths between nodes. - **Path length control**: A maximum path length parameter \( d \) is introduced to ensure that both direct and extended influence are captured simultaneously. - **Standardizing influence**: By dividing the squared degree of a node by the number of paths, the influence of the node is standardized to avoid excessive amplification of highly - connected nodes. ### Formula The CDPs centrality score is defined as follows: \[ \text{CDP}(x)=\frac{\text{deg}(x)^{2}}{\sum_{y \in V}|P(x, y)_{l \leq d}|} \] where: - \( \text{deg}(x) \) represents the degree of node \( x \). - \( |P(x, y)_{l \leq d}| \) represents the number of all paths from node \( x \) to node \( y \), and the path length \( l \) does not exceed the predefined maximum value \( d \). ### Conclusion This paper provides a more comprehensive and effective method for identifying key nodes in social networks by introducing the CDPs method, which makes up for the deficiencies of traditional centrality measurement methods. Experiments have proved that the CDPs method shows significant advantages in a variety of real - world networks.