We present fully dynamic algorithms for maintaining betweenness centrality (BC) of vertices in a directed graph G = (V,E) with positive edge weights. BC is a widely used parameter in the analysis of large complex networks. We achieve an amortized O(ν∗2 · log3 n) time per update with our basic algorithm, and O(ν∗2 · log2 n) time with a more complex algorithm, where n = |V |, and ν∗ bounds the number of distinct edges that lie on shortest paths through any single vertex. For graphs with ν ∗ = O(n), our algorithms match the fully dynamic all pairs shortest paths (APSP) bounds of Demetrescu and Italiano [8] and Thorup [28] for unique shortest paths, where ν∗ = n - 1. Our first algorithm also contains within it, a method and analysis for obtaining fully dynamic APSP from a decremental algorithm, that differs from the one in [8].
CITATION STYLE
Pontecorvi, M., & Ramachandran, V. (2015). Fully dynamic betweenness centrality. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9472, pp. 331–342). Springer Verlag. https://doi.org/10.1007/978-3-662-48971-0_29
Mendeley helps you to discover research relevant for your work.