Interval orders and dimension (Q1970712): Difference between revisions

From MaRDI portal
Added link to MaRDI item.
Import240304020342 (talk | contribs)
Set profile property.
 
(2 intermediate revisions by one other user not shown)
Property / author
 
Property / author: Q200923 / rank
Normal rank
 
Property / author
 
Property / author: William T. jun. Trotter / rank
Normal rank
 
Property / reviewed by
 
Property / reviewed by: Joseph Neggers / rank
Normal rank
 
Property / author
 
Property / author: Henry A. Kierstead / rank
 
Normal rank
Property / author
 
Property / author: William T. jun. Trotter / rank
 
Normal rank
Property / reviewed by
 
Property / reviewed by: Joseph Neggers / rank
 
Normal rank
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Latest revision as of 05:24, 5 March 2024

scientific article
Language Label Description Also known as
English
Interval orders and dimension
scientific article

    Statements

    Interval orders and dimension (English)
    0 references
    21 March 2000
    0 references
    An interval order is a poset whose elements can be interpreted as bounded closed intervals of the real numbers with intervals \(I<J\) if the right endpoint of \(I\) is to the left of the left endpoint of \(J\) on the real line. At the same time using such an interpretation (non unique) also determines an overlap graph (non unique) where \(I\) and \(J\) overlap if they intersect but neither is contained in the other. Interval orders are a special class of posets which have been well studied and for which many special properties hold which do not hold in general. In this paper another such ``class-dependent'' theorem is proven, viz: for every finite interval order \(X\) there exists a positive integer \(t=t(X)\) such that if \(Y\) is any interval order of dimension at least \(t\), then \(Y\) contains a subposet isomorphic to \(X\). In particular, if \(|X|=n\), then \(t=3n-6\) will work. This shows that interval orders are ``sufficiently scarce'' to admit such a theorem. Trivially, chains and antichains are obviously of the same type with \(t(X)=1\) or 2 respectively. The authors suspect that \(t=3n-6\) is not the best estimate.
    0 references
    embedding
    0 references
    dimension
    0 references
    interval order
    0 references
    overlap graph
    0 references

    Identifiers