文章基本信息

标题：Fully Dynamic Connectivity Oracles under General Vertex Updates
本地全文：下载
作者：Kengo Nakamura
期刊名称：LIPIcs : Leibniz International Proceedings in Informatics
电子版ISSN：1868-8969
出版年度：2017
卷号：92
页码：59:1-59:12
DOI：10.4230/LIPIcs.ISAAC.2017.59
出版社：Schloss Dagstuhl -- Leibniz-Zentrum fuer Informatik
摘要：We study the following dynamic graph problem: given an undirected graph G, we maintain a connectivity oracle between any two vertices in G under any on-line sequence of vertex deletions and insertions with incident edges. We propose two algorithms for this problem: an amortized update time deterministic one and a worst case update time Monte Carlo one. Both of them allow an arbitrary number of new vertices to insert. The update time complexity of the former algorithm is no worse than the existing algorithms, which allow only limited number of vertices to insert. Moreover, for relatively dense graphs, we can expect that the update time bound of the former algorithm meets a lower bound, and that of the latter algorithm can be seen as a substantial improvement of the existing result by introducing randomization.
关键词：Dynamic Graph; Connectivity; Depth-First Search