Graph
description
Transcript of Graph
![Page 1: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/1.jpg)
Graph
• graph is a pair of two sets where– set of elements called vertices (singl. vertex)– set of pairs of vertices (elements of ) called edges
• Example: where
• Notation: – we write for a graph with vertex set and edge set – is the vertex set of , and is the edge set of
1
2
3 4
drawing of
![Page 2: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/2.jpg)
1 2 3 4
1 1 1
2 1 1
3 1 1 1
4 1
1 2 3 4
1 1 1
2 1
3 1 1
4
1 2 3 4
1 1 3
2 1 2
3 3 2 4
4 4
1 2 3 4
1 1 1
2 1 1 1
3 1 1 1 1
4 0 1
Types of graphs
• Undirected graph
set of unordered pairs
• Pseudograph (allows loops)
loop = edge between vertex and itself
1
2
3 4
• Directed graph
set of ordered pairs
• Multigraph
is a multiset
(IRREFLEXIVE) RELATIONSYMMETRIC & IRREFLEXIVE
SYMMETRIC ?
1
2
3 4
1
2
3 4
1
2
3 4
![Page 3: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/3.jpg)
• simple graph (undirected, no loops, no parallel edges)for edge we say:– and are adjacent– and are neighbours– and are endpoints of – we write for simplicity• = set of neighbours of in G• = degree of v is the number of its neighbours, ie
How many simple graph on vertices are there?1
23
1
23
More graph terminology
1
2
3 4
1 is adjacent to 2 and 3
1
23
1
23
1
23
1
23
1
23
1
23
1
23¿ ≠
is isomorphic to if there exists a bijective mapping such that if and only if
≅
![Page 4: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/4.jpg)
1
Isomorphism
2 3
4 5 6
4
5
6
1
2
3
4
6
5
1
3
2
(3,3,3,3,3,3) (3,3,3,3,3,3) (3,3,3,3,3,3)
(3,3,3,3,2,2,2,2) (3,3,3,3,2,2,2,2)
3 3 3
3 3 3
≅
if is an isomorphism
![Page 5: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/5.jpg)
Isomorphism
How many pairwise non-isomorphic graphs on vertices are there?
the complement of is the graph where
1
23
1
23
1
23
1
23
1
23
1
23
1
23
1
23
2
1
3
4
5
𝐺
2
1
3
4
5
𝐺
, ≅ self-complementary graph
![Page 6: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/6.jpg)
Isomorphism
• Are there self complementary graphs on 5 vertices ?
• ... 6 vertices ? No, because in a self-complementary graph and but is odd• ... 7 vertices ?• ... 8 vertices ?
2
1
3
4
5
2
1
3
4
5
4
1
5
8 7
6
2
3
No, since
Yes, there are 10
Yes, 2
![Page 7: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/7.jpg)
Handshake lemma
Lemma. Let be a graph with edges.
Proof. Every edge connects 2 vertices and contributes to exactly 1 to and exactly 1 to . In other words, in every edge is counted twice.
How many edges has a graph with degree sequence:- (3,3,3,3,3,3) ?- (3,3,3,3,3) ?- (0,1,2,3) ?
Corollary. In any graph, the number of vertices of odd degree is even.Corollary 2. Every graph with at least 2 vertices has 2 vertices of the same degree.
4
5
6
1
2
3
1
2
3 4
𝑚=4
Is it possible for a self-complementary graph with 100 vertices to have exactly one vertex of degree 50 ?
Answer. NoN (𝑢)50
𝑉 (𝐺)¿(𝑢)49
N 𝐺(𝑢)49
isomorphism between and
is a unique vertex of degree in
the degree seq. of and
𝑣𝑓−1
∃𝑣 , 𝑓 (𝑣 )=𝑢
we conclude
𝑓𝑢
𝑢
definition of isomorphism definition of complement
but
![Page 8: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/8.jpg)
Subgraphs, Paths and Cycles
Let be a graph• a subgraph of is a graph where and • an induced subgraph (a subgraph induced on ) is a graph
where • a walk (of length ) in is a sequence
where and for all • a path in is a walk where are distinct
(does not go through the same vertex twice)• a closed walk in is a walk with
(starts and ends in the same vertex)• a cycle in is a closed walk where and are distinct
1
2
3 4
1
2
3 4
2
3 4
subgraph induced subgraphon 1
2
3 4
1,{1,2},2,{2,1},1,{1,3},3is a walk (not path)
1,{1,2},2,{2,3},3,{3,1},1is a cycle
![Page 9: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/9.jpg)
Special graphs
• path on vertices• cycle on vertices• complete graph on vertices• hypercube of dimension
adjacent to if ’s and ’s differ in exactly once
adjacent to not adjacent to
∑𝑖=1
𝑛
|𝑎𝑖−𝑏𝑖|=1
2
1
3
4
5
21 3 4 5 𝑃5
𝐶5
2
1
3
4
5
𝐾 5
𝐵3
000
100010
001
110 111
011101
![Page 10: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/10.jpg)
Connectivity
• a graph is connected if for any two vertices of there exists a path (walk) in starting in and ending in
• a connected component of is a maximal connected subgraph of
1
2
3 4
1
2
3 4
connected not connected = disconnected
no path from 1 to 4
connected components1,{1,3},3,{3,4},4path from 1 to 41
3 4
![Page 11: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/11.jpg)
Connectivity• Show that the complement of a disconnected graph is connected !
• What is the maximum number of edges in a disconnected graph ?
• What is the minimum number of edges in a connected graph ?
𝑢𝑣
𝑢𝑣
𝐺 𝐺
(𝑘2 ) (𝑛−𝑘2 ) max𝑘 (𝑘2 )+(𝑛−𝑘2 )=(𝑛−12 )
maximum when
21 3 4 5
path on vertices has edges
cannot be less, why ? keep removing edges so long as you keep the graph connected a (spanning) tree which has edges
![Page 12: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/12.jpg)
Distance, Diameter and Radius
• length of a walk (path) is the number of edges• distance from a vertex to a vertex is the length of a shortest
path in between and • if no path exists define • eccentricity of a vertex is the largest distance between and
any other vertex of ; we write • diameter of is the largest distance between two vertices
• radius of is the minimum eccentricity of a vertex of
Find radius and diameter of graphs , , Prove that
recall walk (path) is a sequence of vertices and edges
if edges are clear from context we write
1
2
3 4
1
2
3 4(as witnessed by 3 called a centre)
(as witnessed by 1,4called a diametrical pair)
![Page 13: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/13.jpg)
Independent set and Clique
• clique = set of pairwise adjacent vertices• independent set = set of pairwise non-adjacent vertices• clique number = size of largest clique in • independence number = size of largest indepen-dent set
in
Find , , , ,
1
2
3 4
is a clique
is an independent set
21 3 4 5
𝑃5
![Page 14: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/14.jpg)
Bipartite graph
• a graph is bipartite if its vertex set can be partitioned into two independent sets , ; ie.,
1
2
3 41
2
3 4
1 2
34
bipartite not bipartite
1∈𝑉 1
2∈𝑉 2
3∈𝑉 2but now is notan independent set because
𝑉 1
𝑉 2
𝑉 2𝑉 1
𝐾 2,2
𝐾 3,3
𝐾 4,4 = complete bipartite graph
Which of thesegraphs are bipartite:
, , ?
𝐾 1,2
![Page 15: Graph](https://reader038.fdocuments.in/reader038/viewer/2022103007/568143a7550346895db02db4/html5/thumbnails/15.jpg)
Bipartite graphTheorem. A graph is bipartite it has no cycle of odd length.
Proof. Let . We may assume that is connected.“” any cycle in a bipartite graph is of even length
“” Assume G has no odd-length cycle. Fix a vertex and put it in . Then repeat as long as possible:- take any vertex in and put its neighbours in , or- take any vertex in and put its neighbours in .Afterwards because is connected.If one of or is not an independent set, then we find in an odd-length closed walk cycle, impossible.So, is bipartite (as certified by the partition )
2
1
3 45
𝑘...
2
1
3
2
1
3
4
5
2
1
34
56
2
3
4
1
𝑢
?
odd