Compression and Erdős-Ko-Rado graphs

From MaRDI portal





This paper deals with the following, rather general problem. Let \(G\) be a graph, \({\mathcal I}^{(r)}(G)\) denote its \(r\)-element independent sets, while \({\mathcal I}^{(r)}_v(G)\) denote those \(r\)-element independent sets which contain vertex \(v.\) A graph is called \(r\)-EKR if there is no bigger intersecting subfamily in \({\mathcal I}^{(r)}(G)\) than the maximum size \(| {\mathcal I}^{(r)}_v(G)| .\) The main question is to describe the \(r\)-EKR graphs. If \(G\) is an empty graph with \(n\) vertices, then to determine the biggest \(r\) such that \(G\) is \(r\)-EKR is the well-known Erdős-Ko-Rado problem. This paper, using a compression technique, shows that the disjoint union of at least \(r\) complete graphs (with order of at least 2 each) is \(r\)-EKR. It is also shown that the paths are \(r\)-EKR.




Cited in
(41)








This page was built for publication: Compression and Erdős-Ko-Rado graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1779494)