Graph classes between parity and distance-hereditary graphs (Q1302157)

From MaRDI portal
Revision as of 08:32, 30 July 2024 by Openalex240730090724 (talk | contribs) (Set OpenAlex properties.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
Graph classes between parity and distance-hereditary graphs
scientific article

    Statements

    Graph classes between parity and distance-hereditary graphs (English)
    0 references
    0 references
    0 references
    22 March 2000
    0 references
    We define operations \(\beta\), \(\gamma\) and \(\phi\) as follows: Let \(x\) be a vertex of a graph \(G\) and let \(y\) be a vertex of a graph \(H\) such that \(G=H-y\). Then \(H = \beta(G,x)\) if \(x\) and \(y\) are true twins (both vertices have the same closed neighbourhood in \(H\)), and \(H = \gamma(G,x)\) if \(x\) and \(y\) are false twins (both vertices have the same open neighbourhood in \(H\)). Now let \((X,Y,Z)\) be a partition of the vertex set of \(H\) such that \(G = H-Y\). Then \(H = \phi(G,B,X)\) if \(X\) is a set of false twins in \(G\) and \(X\) is contained in one color class of the bipartite graph \(B = H-Z\). Let \(C\) be a class of bipartite graphs closed under taking connected subgraphs. By \(\Phi_C\) we denote the hull of \(K_1\) under the operations \(\beta\), \(\gamma\) and \(\phi\), where \(B \in C\). If \(C\) is the class of all bipartite graphs then \(\Phi_C\) is the class of all parity graphs, and if \(C = \{K_2,K_1\}\) then \(\Phi_C\) is the class of all distance hereditary graphs. In general, the classes \(\Phi_C\), partially ordered by set inclusion, form a lattice of infinite height. Cunningham's split decomposition leads to characterizations of the \(\Phi_C\) classes that enable polynomial time algorithms for recognition and isomorphism test.
    0 references
    parity graph
    0 references
    distance-hereditary graph
    0 references
    split decomposition
    0 references
    recognition algorithm
    0 references
    graph isomorphism
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references