directed graph

directed graph
Name	Directed graph
Caption	A small directed graph with five vertices and six edges
Type	Mathematical object
Fields	Graph theory, Combinatorics, Computer science
Introduced	18th–20th centuries

directed graph A directed graph is a mathematical structure consisting of a set of vertices connected by edges that have an orientation. It appears across Leonhard Euler's work on the Seven Bridges of Königsberg, the development of Gustave de Rebeur-Paschwitz's and Arthur Cayley's discrete studies, and modern computational treatments in Donald Knuth's volumes and research at institutions such as Bell Labs and MIT. Directed graphs underpin models in theoretical work by Paul Erdős, applications in systems studied at Stanford University and Carnegie Mellon University, and algorithmic advances by researchers at IBM and Microsoft Research.

Definition and basic concepts

A directed graph consists of a finite or countable set of vertices together with a set of ordered pairs of vertices called arcs or directed edges; classical expositions appear in texts by Frank Harary, Claude Berge, William Tutte, and Reinhard Diestel. Core notions include in-degree and out-degree of a vertex, sources and sinks, loops (edges from a vertex to itself), and parallel arcs; foundational examples and counterexamples are studied in seminars at Princeton University and University of Cambridge. Related constructs like tournaments, directed acyclic graphs, and multigraphs are treated in curriculum at University of Oxford and research at ETH Zurich.

Properties and types

Important structural properties studied by Eugène Wigner-era combinatorialists and later by László Lovász include connectivity (strong and weak), reachability, acyclicity, and planarity constraints in orientations of undirected graphs as explored in work associated with Havel and Hakimi. Special classes include tournaments (complete asymmetric orientations studied by Paul Erdős and Joel Spencer), oriented trees and arborescences (investigated in papers by József Kőnig and Gabriel Andrew Dirac), Eulerian and Hamiltonian orientations (with results by Leonhard Euler and conjectures related to William Rowan Hamilton), and planar orientations relevant to studies at Institut des Hautes Études Scientifiques and Max Planck Institute for Mathematics. Spectral properties of adjacency and Laplacian matrices of directed graphs have been advanced by researchers at Princeton University and ETH Zurich.

Representations and data structures

Representations include adjacency matrices, adjacency lists, incidence matrices, and compact dynamic structures used in systems built at Google and Facebook. Adjacency matrices facilitate algebraic techniques used in work by Alfred North Whitehead-inspired algebraists and in numerical analysis at Los Alamos National Laboratory, while adjacency lists and edge lists are preferred in large-scale graph processing frameworks developed at Yahoo! Research and Twitter Research. Compressed sparse row formats, succinct encodings, and external-memory representations arise in implementations at Intel and in algorithms courses at University of California, Berkeley.

Algorithms and computations

Key algorithms include depth-first search and breadth-first search popularized in textbooks by Robert Tarjan and Thomas H. Cormen, shortest-path algorithms like Bellman–Ford and Dijkstra associated with work at Bell Labs and AT&T, topological sorting for acyclic orientations studied in research by Edsger W. Dijkstra and Niklaus Wirth, and strongly connected components algorithms by Kosaraju and Tarjan. Flow algorithms such as Edmonds–Karp and push–relabel stem from combinatorial optimization research at Stanford University and Rutgers University, while matching and network design draw on contributions by Jack Edmonds and John Hopcroft. Parallel and distributed algorithms for massive directed graphs have been advanced in projects at Google, Microsoft Research, and Facebook AI Research.

Applications

Directed graphs model dependencies in compilers and build systems used at Bell Labs and AT&T Labs, citation and hyperlink structures analyzed by Google's PageRank team and by scholars at Cornell University, social networks engineered at Facebook and Twitter, transportation and traffic flow analyses in studies by MIT and Imperial College London, and biological regulatory networks researched at Cold Spring Harbor Laboratory and Harvard Medical School. They also underpin scheduling and precedence constraints in projects at Boeing and Siemens, information-flow models in cybersecurity programs at DARPA, and state-transition systems used in formal verification at Microsoft Research and IBM Research.

History and notation conventions

Origins trace to Leonhard Euler's 1736 solution of the Seven Bridges of Königsberg problem and to 19th- and 20th-century formalizations by Arthur Cayley, James Joseph Sylvester, and later systematization by Frank Harary and Claude Berge. Notation conventions—use of ordered pairs for arcs, matrix notation for adjacency, and terms like in-degree and out-degree—were standardized in textbooks by Reinhard Diestel and monographs from Springer-Verlag and Cambridge University Press, with computational conventions adopted in software libraries at Boost and GNU. Developments in the late 20th century at institutions including Bell Labs, MIT, and Stanford University integrated algebraic, combinatorial, and algorithmic perspectives that continue to evolve in contemporary research at Simons Foundation-funded centers and national laboratories.

Category:Graph theory