Eccentricity of networks with structural constraints (Q2194537)

This paper studies eccentricity centrality for bipartite graphs and tree graphs with structural restrictions. Eccentricity centrality \(E(v)=1/e(v)\) is defined as the reciprocal of the eccentricity of a node \(v\), and \(e(v)\) is the maximum distance between \(v\) and all nodes \(u\) in a connected graph \(G\). Define \(E_1(v)=\sum_{u\in V(G)}(E(v)-E(u))\). For a connected graph with bipartite sets \(K\) and \(L\) of sizes \(|K|=k\ge2\) and \(|L|=l\ge2\), it is shown \(E_1(v)\le l/6\) for \(k=2\) and \(v\in K\). Moreover, if \(k\ge3\), then \(E_1(v)\le (l/6)+(k-1)/4\). On estimating the eccentricity centrality of trees with prescribed order and maximum degree, a scheme dubbed as \(S\)-enumerations has been developed to label the vertices of a tree in the characterization of trees with maximum eccentricity.