Node local centrality (NLC)
The
Node local centrality
(NLC) is a centrality measure for identifying influential spreaders by combining network embedding (NE) with local network information [2].
For a node \(i\), the NLC centrality \(c_{\textsc{NLC}}(i)\) is defined as
\[
c_{\textsc{NLC}}(i) = \sum_{j \in \mathcal{N}(i)} k_s(i) \, e^{-\|x_i - x_j\|^2},
\]
where \(k_s(i)\) is the \(k\)-shell index of node \(i\) and \(x_i \in \mathbb{R}^{r \times 1}\) is the vector embedding of node \(i\) obtained via the DeepWalk network representation method [3].
Yang et al. [2] set the embedding dimension to \(r = N/2\). The measure captures both the hierarchical position of a node in the network (via \(k\)-shell) and its proximity in the embedded low-dimensional space, reflecting local structural similarity.