IV
IVI Interactive Demo
Integrated Value of Influence

IVI is the first integrative method that captures all topological dimensions of a network. It synergizes six centrality measures while removing their inherent biases to identify the most influential nodes.

🔬 Live Network with IVI Coloring
High IVI
Medium IVI
Low IVI
Uncomputed
🌟
Local Centrality
Immediate neighborhood

Measures how well-connected a node is to its direct neighbors. IVI integrates Degree Centrality and ClusterRank to capture local influence while correcting for clustering bias.

🔗
Semi-Local Centrality
Second-order neighborhood

Captures influence beyond immediate neighbors. IVI integrates Neighborhood Connectivity and Local H-index to assess a node's power within its extended local environment.

🌎
Global Centrality
Whole-network position

Measures a node's position in the entire network topology. IVI integrates Betweenness Centrality and Collective Influence to capture global bridging and spreading power.

📊
The IVI Formula
Synergy + bias removal

IVI = Hubness Score × Spreading Score. Hubness combines DC and LH-index; Spreading combines CR, NC, BC, and CI. This multiplicative integration ensures both local power and global reach.

📈 IVI vs Other Methods
Method Local Semi-Local Global Bias Correction
Degree Centrality
Betweenness
PageRank ~
K-shell ~ ~
★ IVI
🔄 How IVI Works
1
Network Input
Provide an igraph object (directed/undirected, weighted/unweighted)
2
Compute Six Centralities
DC, CR, NC, LH-index, BC, and CI are calculated for every node
3
Integrate via IVI Formula
Hubness Score × Spreading Score = IVI (range-normalized 0-100)
4
Rank & Visualize
Use cent_network.vis() to color and size nodes by IVI
📄 Salavaty A, Ramialison M, Currie PD. Integrated Value of Influence: An Integrative Method for the Identification of the Most Influential Nodes within Networks. Patterns. 2020.
💻 Available in R/Python via the influential package