Simple Graph
A graph with no loops and no multiple edges between the same pair of vertices.
Pseudograph
Aka. multi-graph. Consists of a set of vertices (aka. nodes or points) and some pair of vertices are connected by edges (aka. line segments).
The set of vertices is usually denoted by . The set of edges is denoted by . The pseudograph is denoted as .
Regular Graph
A graph where all vertices have the same degree (). Aka. -regular graph.
Null Graph
A graph with no edges. All vertices are isolated.
Directed Graph
Aka. digraph. Edges have direction.
where is a set of ordered pairs of vertices. Edges of a directed graph are called arcs or directed edges or arrows or directed lines.
Starting vertex (where the arrow starts from) is called the tail, ending vertex (where the arrow points to) is called the head. Vertices have in-degree (denoted as ) and out-degree (denoted as ) which means the number of incoming and outgoing edges respectively. Loops contribute 1 to both in-degree and out-degree.
Simple and multi digraph versions exist.
Complete Graph
A simple graph where every pair of vertices is adjacent. Denoted by where is the number of vertices.
Contains number of edges, which is the maximum for a simple graph with vertices.
Bipartite Graph
A graph whose vertices can be split into disjoint sets and such that:
- All edges go between and ,
- No edges exist inside or inside .
Here and are called bipartition sets.
Complete Bipartite Graph
Every vertex in is connected to every vertex in . Denoted by .
Self-Complementary Graph
A graph satisfying .
If, is self-complementary, has edges.
Cyclic Graph
A graph with at least 1 cycle.
A simple graph consisting of only a cycle with vertices, is called -cyclic graph and denoted as .
is only possible as a multi-graph. Otherwise it will be a open path and not a cycle.
Acyclic Graph
A graph with no cycles.
Euler Graph
Aka. Eulerian graph. A connected graph that contains an Eulerian circuit.
A graph is Eulerian iff every vertex has an even degree. Whenever a walk enters a vertex, it must leave via another edge. Thus edges are used in pairs, making the vertex degree even.
Examples:
- (2 vertices connected by 2 parallel edges)
The smallest Eulerian multi graph. - (triangle)
The smallest Eulerian simple graph.
Hamiltonian Graph
A graph that contains a Hamiltonian cycle. Introduced by William Hamilton in 1857, while studying cycles in the dodecahedron graph.
Dirac’s Theorem
A simple graph with vertices is Hamiltonian if every vertex has degree at least .
Ore’s Theorem
A simple graph with vertices is Hamiltonian if for every pair of non-adjacent vertices and :
Weighted Graph
A graph in which each edge is assigned a numerical value (aka. weight). 3-tuples, where the third element is a weight function, .
Weights often represent distance, cost, time, or capacity. Used in road networks, computer network routing, project scheduling, etc.
Planar Graphs
A graph that can be drawn on a plane without any edges crossing.
A drawing of the graph without edge crossings is called a planar representation or planar embedding. A graph can have multiple planar representations.
Euler’s Formula for Planar Graphs
For a connected planar graph:
Here:
- – number of vertices
- – number of edges
- – number of regions (faces), including the outer region.
Corollary of Euler’s Formula
For a connected planar simple graph with , the graph must satisfy .
If a graph violates this inequality, it cannot be planar.
Non-Planar Graph
Certain graphs cannot be drawn without edge crossings.
Examples:
Complete graph with 5 vertices. Does not satisfy the corollary of Euler’s formula for planar graphs, thus non-planar.
Complete bipartite graph with partitions of size 3. Satisfies the corollary of Euler’s formula, but is still non-planar. It can be shown that it contains a subgraph homeomorphic to .
These are the smallest non-planar graphs.
Kuratowski’s Theorem
A graph is non-planar iff it contains a subgraph homeomorphic to or .