Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data. A simulation from public publication records: the people listed were not invited and did not agree to review.
? Tutorial for this page

André E. Kézdy

dblp:11/1817 · DBLP profile ↗
← Back
9ranked-venue papers
2as first author
2since 2021 · last 2026
0000-0002-5389-6758ORCID · verified

Domains — the database's venue-derived domains; a paper can count in several

Theory of computation · 7 · 2 first-author · 2 since 2021Computer networks · 2Databases, data management, data science and information retrieval · 2 · 1 since 2021

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Theoretical computer science
1 paper
Graph algorithms and graph theory · 50% Algorithms and data structures · 50%

Topics — the 2 heaviest of 2, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Graph algorithms and graph theory
graph minors
0.011992
Sequential and Parallel Algorithms to Find a K5 Minor · SODA 1992
Algorithms and data structures
parallel algorithms
0.011992
Sequential and Parallel Algorithms to Find a K5 Minor · SODA 1992
YearPublicationVenuePosition
2026 The Schrijver system of the length polyhedron of an interval order
André E. Kézdy, Jenö Lehel
Discret. Appl. Math.1
2021 Minimal 2-connected graphs satisfying the even cut condition
Adam S. Jobson, André E. Kézdy, Jenö Lehel
Inf. Process. Lett.2
2020 Note on the bisection width of cubic graphs
abstract
The bisection width is the minimum number of edges required to split the vertex set of a graph into two (nearly) equal parts. Monien and Preis proved that the bisection width of a cubic graph with n nodes is bounded above by n∕6+o(n). Here we show that every cubic graph of even order n≥16 has bisection width less than n∕2, thus these graphs violate the even cut condition (ECC). All edge-minimal subcubic graphs satisfying ECC are also described. The bisection width is a reference parameter to compare networks for parallel architectures; ECC is a property necessary for bottleneck free all-to-all communications.
Adam S. Jobson, André E. Kézdy, Jenö Lehel
Discret. Appl. Math.2
2018 The minimum size of graphs satisfying cut conditions
Adam S. Jobson, André E. Kézdy, Jenö Lehel
Discret. Appl. Math.2
2018 Linkage on the infinite grid
Adam S. Jobson, André E. Kézdy, Jenö Lehel
Inf. Process. Lett.2
2016 Detour trees
Adam S. Jobson, André E. Kézdy, Jenö Lehel, Susan C. White
Discret. Appl. Math.2
1998 Tough enough chordal graphs are Hamiltonian
abstract
We prove that every 18-tough chordal graph has a Hamiltonian cycle. © 1998 John Wiley & Sons, Inc. Networks 31: 29–38, 1998
Guantao Chen, Michael S. Jacobson, André E. Kézdy, Jenö Lehel
Networks3
1998 Recognizing triangle-free graphs with induced path-cycle double covers is NP-complete
abstract
An induced path-cycle double cover (IPCDC) of a simple graph G is a family ℱ = {F1, …, Fk} of induced paths and cycles of G such that if Fi ∩ Fj ≠ ⊘, then Fi ∩ Fj is a vertex or an edge, for i ≠ j, each edge of G appears in precisely two of the Fi's, and each vertex of G appears in precisely three of the Fi's. In this paper, we prove that recognizing triangle-free simple graphs with an IPCDC is NP-complete by reducing the 3-Satisfiability problem to finding an IPCDC. The dependency graph of a 3-uniform hypergraph H = (V, E) is the graph with vertex set E in which two hyperedges are joined if and only if they share exactly two elements. We show that a triangle-free simple graph is the dependency graph of a 3-uniform hypergraph if and only if it has an IPCDC. Consequently, the problem of recognizing dependency graphs of 3-uniform hypergraphs is NP-complete. © 1998 John Wiley & Sons, Inc. Networks 31: 1–10, 1998
Michael S. Jacobson, André E. Kézdy, Jenö Lehel
Networks2
1992 Sequential and Parallel Algorithms to Find a K5 Minor
André E. Kézdy, Patrick McGuinness
SODA1