Efficient algorithms for measuring the funnel-likeness of DAGs

Marcelo Garlet Millani, Hendrik Molter, Rolf Niedermeier, Manuel Sorge

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations


Funnels are a new natural subclass of DAGs. Intuitively, a DAG is a funnel if every source-sink path can be uniquely identified by one of its arcs. Funnels are an analog to trees for directed graphs that is more restrictive than DAGs but more expressive than in-/out-trees. Computational problems such as finding vertex-disjoint paths or tracking the origin of memes remain NP-hard on DAGs while on funnels they become solvable in polynomial time. Our main focus is the algorithmic complexity of finding out how funnel-like a given DAG is. To this end, we study the NP-hard problem of computing the arc-deletion distance to a funnel of a given DAG. We develop efficient exact and approximation algorithms for the problem and test them on synthetic random graphs and real-world graphs.

Original languageEnglish
Title of host publicationCombinatorial Optimization - 5th International Symposium, ISCO 2018, Revised Selected Papers
EditorsGiovanni Rinaldi, A. Ridha Mahjoub, Jon Lee
PublisherSpringer Verlag
Number of pages13
ISBN (Print)9783319961507
StatePublished - 1 Jan 2018
Event5th International Symposium on Combinatorial Optimization, ISCO 2018 - Marrakesh, Morocco
Duration: 11 Apr 201813 Apr 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10856 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference5th International Symposium on Combinatorial Optimization, ISCO 2018

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Efficient algorithms for measuring the funnel-likeness of DAGs'. Together they form a unique fingerprint.

Cite this