UNIVERSITY OF STIRLING MATPMD2

Computing Science & Mathematics 2020

MATPMD2 : NETWORKS and GRAPH THEORY

1 Graph Theory
Graph theory is about connections between objects. Informally a graph is a set of vertices or nodes,
together with a set of edges or lines between pairs of vertices. The diagram illustrating this
information is commonly referred to as a graph.

Example 1.1 Bus routes.

Here the objects are villages and the connections are bus routes between them. Multiple edges arise
because of multiple service providers; the loop is a scenic tour. If we are interested solely in whether
or not you can get from one village to another by bus then we can disregard the multiple edges, and
the loop.

Simplified:
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

Example 1.2 Phone calls.

Here the objects are people and the connections are phone calls made from one person to another. The
edges are given a direction and we call this a directed graph, or digraph.

The number of edges incident with a vertex is called the degree of a vertex. For a digraph this idea is
extended to in-degrees (the number of edges going into a vertex) and out-degrees (the number of
edges leaving a vertex). An edge going in both directions can be represented by an undirected edge.

Example 1.3 Molecular structure.

Here the objects are atoms and the connections are molecular bonds.
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

Example 1.4 Incidence graph.

Here the objects can be described as points p1 , p2 , . . . , pk and sets of points S1 , S2 , . . . Sr . The
connections arise between the points and the sets of points with an edge between pi and Sj if and only
if pi ∈ Sj .

One application is suitability matching; we construct the corresponding incidence graph and obtain a
perfect matching (a set of pairwise disjoint edges) if possible.
In this example we have five people and five jobs. Here Si is the set of jobs person i is suited to:

S1 = {a, d}, S2 = {a, c, d}, S3 = {e}, S4 = {a.c}, S5 = {a, b}.

Perfect matching:

The perfect matching shows person 1 should do job d, person 2 job c, person 3 job e, person 4 job a
and person 5 job b.
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

1.1 Some definitions and some graphs.

A graph G consists of a vertex set V (G) and an edge set E(G) where E(G) is a subset of V (G)2 . We
confine ourselves to simple graphs; undirected edges with no loops or multiple edges but denote an
edge by (a, b). If (a, b) ∈ E(G) we say vertex a is adjacent to vertex b and write a ∼ b. For a given
vertex v, the number of edges incident with v is the degree of v; vertices adjacent to v are called the
neighbours of v.

Example

Let V (G) = V = {1, 2, 3, 4, 5}, E(G) = {(1, 2), (1, 3), (1, 4), (2, 5), (3, 4)}.
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

Given a graph G, the complement G has vertex set V (G) = V (G) and edge set
E(G) = V (G)2 \ E(G).
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

Given a graph G, a subgraph H of G is such that V (H) ⊆ V (G) and E(H) ⊆ E(G).

If we fix a vertex v in a graph G, then G − v is the vertex-deleted subgraph of G obtained by deleting

the vertex v. Note that deletion of a vertex means deletion of any edge containing that vertex, or em
incident with that vertex. Deletion of an edge does not mean you need to delete vertices in that edge.
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

n(n−1)
The complete graph Kn has n vertices and all possible 2
edges. The complement of Kn is n
isolated vertices, denoted by nK1 .

A k-regular graph is a connected graph for which each vertex has degree k.

The Petersen Graph:

A cycle is a connected graph which is regular of degree two.

UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

A complete bipartite graph has a vertex set which can be partitioned into two sets of isolated vertices
and an edge set consisting of all the edges between the two vertex sets. A bipartite graph is any
subgraph of a complete bipartite graph. The incidence graph is a bipartite graph.

All bipartite graphs have a 2-colouring. This means that we need no more than two colours to colour
vertices so that adjacent vertices have different colours.
UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

A walk is a sequence of vertices; called a trail if the edges are distinct, a path if the vertices are
distinct. A graph is connected if there exists a walk between every pair of vertices. A graph which is
not connected has two or more components.

A u − v walk is a sequence of vertices v1 v2 v3 . . . vr where u = v1 and v = vr .

A tree is a connected graph with no cycles; a forest is a graph with no cycles.

A tree (known for its bark)

and a forest of two trees (for her to play in.)

UNIVERSITY OF STIRLING MATPMD2
Computing Science & Mathematics 2020

The incidence matrix of a graph G is a matrix whose rows are indexed by V (G) and whose columns
are indexed by E(G) ; the (v, λ)-entry is 1 if v ∈ λ and 0 if v 6∈ λ. [Here v is a vertex, λ an edge.]

Each row sum gives the degree of the vertex.

Each column sum is 2 since we have two vertices for each edge.

By considering the incidenceP

matrix we have:
Lemma 1 (Euler’s theorem) ni=1 deg(vi ) = 2m where m is the number of edges.