Transcript Digraphs
Directed Graphs
3/6/12
1
Normal Person’s Graph
y
y = f(x)
x
3/6/12
2
Computer Scientist’s Graph
a
b
c
d
f
e
3/6/12
3
Digraphs
• a set, V, of vertices
aka “nodes”
• a set, E V×V
of directed edges
(v,w) E notation: vw
v
3/6/12
w
4
Relations and Graphs
a
d
b
c
V= {a,b,c,d}
E = {(a,b), (a,c), (c,b)}
3/6/12
5
Digraphs
Formally, a digraph
with vertices V is the
same as a binary
relation on V.
3/6/12
6
Walks & Paths
Walk: follow successive edges
length: 5 edges
(not the 6 vertices)
3/6/12
7
Walks & Paths
Path: walk thru vertices
without repeat vertex
length: 4 edges
3/6/12
8
Walks & Paths
Lemma:
The shortest walk between
two vertices is a path!
Proof: (by contradiction) suppose
path from u to v crossed itself:
c
u
3/6/12
v
9
Walks & Paths
Lemma:
The shortest walk between
two vertices is a path!
Proof:
(by contradiction)
then path
without c---csuppose
is
path
from u to v crossed itself:
shorter!
c
u
3/6/12
v
10
Walks & Paths
Digraph G defines walk
+
relation G
+
u G v iff ∃walk u to v
(the positive walk relation)
“+” means 1 or more
3/6/12
11
Walks & Paths
Digraph G defines walk
*
relation G
*
u G v iff 1 w
u
to
v
alk
2 3
length 0
(the walk relation)
“*” means “0 or more”
3/6/12
12
Cycles
A cycle is a walk whose
only repeat vertex is its
start & end.
(a single vertex is a
length 0 cycle)
3/6/12
13
Cycles
…
v0
v1
v2
vn-1
v0
vi
v0
3/6/12
vi+1
14
Closed Walks & Cycles
Closed walk starts & ends at the
same vertex.
Lemma: The shortest positive
length closed walk containing a
vertex is a positive length cycle!
Proof: similar
3/6/12
15
Directed Acyclic Graph
DAG
has no positive
length cycle
3/6/12
16
lec 7M.16
Directed Acyclic Graph
examples:
DAG
< relation on integers
⊊ relation on sets
prerequisite on classes
3/6/12
17
Example: Tournament Graph
• Every team plays every other
H
H
Y
P
D
Y
P
D
DAG => Unique ranking
3/6/12
18