Scribd
Upload
207 views

Concepts of Graph Theory: Social Networks Lecture 2

This document provides an overview of key concepts in graph theory and social network analysis. It discusses how social networks can be represented as graphs with nodes and edges. It also describes common network measures like degree distribution, density, components, walks/trails/paths, cutpoints/bridges, and centralization. Different types of ties between nodes are explained as well as how networks can be represented using adjacency matrices. Examples are provided to illustrate concepts like the seven bridges of Königsberg problem.

Uploaded by

Maksim Tsvetovat
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
207 views

Concepts of Graph Theory: Social Networks Lecture 2

This document provides an overview of key concepts in graph theory and social network analysis. It discusses how social networks can be represented as graphs with nodes and edges. It also describes common network measures like degree distribution, density, components, walks/trails/paths, cutpoints/bridges, and centralization. Different types of ties between nodes are explained as well as how networks can be represented using adjacency matrices. Examples are provided to illustrate concepts like the seven bridges of Königsberg problem.

Uploaded by

Maksim Tsvetovat
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 26

Concepts of Graph

Theory
Social Networks; Lecture 2
Summary

• Graph representation of social networks

• Matrix representation of social networks

• Node degree; average degree; degree distribution

• Graph density

• Walks, trails and paths

• Cutpoits, cutsets and bridges


What is a Network?

• A set of dyadic ties, all of the same type,among a


set of actors
• Actors can be persons, organizations ...
• A tie is an instance of a social relation
Relations Among Persons

• Kinship
– Mother of, father of, sibling of
• Role-Based
– Boss of, teacher of
– Friend Of
• Affective
– Likes, trusts
• Interactions
– Gives advice to; talks to; sexual interactions
• Affiliations
Content and Coding Matter!

• Each relation yields a different structure


and has different effects
• In real data, more then one relation
should be studied.

• Coding:
– What constitutes an edge?
– How to convert interview data into graph data?
Example
Problem Reformulation
Graph Theoretic Concepts

• Consists of a collection of nodes and


lines
G = N, L
N={n1 , n2 , n3 ...ng }
L = {l1 , l2 , l3 ...lL }

• Lines also called “ties” or “edges”


• Nodes occasionally called “agents” or
“actors”
Directed and Undirected
Ties
• •Undirected relations
Attended meeting with...
• Communicated with...
• Friend of...
• •Directed relations
Represent flows or subordination
• “Lends money to”, “teacher Of”
• •Problem -
Ties that should be symmetric can be measured as non-
symmetric due to measurement error
• Friendship relations are not always reciprocal
Tie Strength

• We can attach values to ties, representing


quantitative attributes
• Strength of relationship
• Frequency of communication
• Information capacity/bandwidth
• Physical distance
• Such graph is called “weighted graph”
4
Adjacency Matrices
!"#$%&'%()*$+,-%&.
!"#$%&'(#)
*#+ *#,, *$% *-$
*#+ . / 0 /
*#,, / . / 0 *#,,
*$% 0 / . / / 5
*-$ / 0 / . *$%
6
*#+

1"-2#+#34 /8
*#+ *#,, *$% *-$ 5
7
*#+ . 5 6 7 *-$
*#,, 5 . / /8
*$% 6 / . 5
*-$ 7 /8 5 .
Sparse Matrix
!"#$%&'%()*$+,-%&.
!"#$%&'(#)
*#+ *#,, *$% *-$
*#+
Jen. Jill/ 0 1/
*#,, / . / 0 *#,,
Jen Joe 3
*$% 0 / . / / 5
Jill Jen
*-$ / 0 /
1. *$%
*#+
Jill Jim 3 6

Jim Jill 3
1"-2#+#34 /8
Jim Joe
*#+ *#,, *$%2*-$ 5
7
Joe. Jim
*#+ 5 6 27 *-$
*#,,
Joe5 Jen . / 3/8
*$% 6 / . 5
*-$ 7 /8 5 .
Node Degree
• Degree of a node is a number of lines that
connect it to other nodes
• Degree can be interpreted as
• measure of power or importance of a node
• or
• measure of workload
• In directed graphs:
• indegree: number of incoming edges
• outdegree: number of outgoing edges
Marriage Ties Among
!"#$%&'
()*#*+',-(./
!"##$"%&'($&)'*+,-%
Leading Florentine Families
./,#&-0$-&'."+$/$&)
1:#$-%';&-"$))"-2&
0$+&)

1"0"'2,+3$/&4'56'7,8-'9"4%&00
Degree Distribution
Graph Density

• Defined as ratio of number of edges in the


graph to the total POSSIBLE number of
edges:
L 2L
∆= =
g(g − 1)/2 g(g − 1)
Density and Network Survival:
Help with rice harvest
!"#$%&'()%()"%*'+"%!,-."/
!"#$%&'()%()"%*'+"%!,-."/(

0,.."1/$2 4#&&'()+5
!"#"$%&'($)*#+,-#./$/#$". 6'1'+7.-,
Components
!"#$%&'()"&*%+","-'./'#$#%0
BC,&4,)&1,&%,&+,&%C'%&4,)&#'$&+'4&DE&0'$&*%&F4&GGGG9&'$/&+C"&+'4+&AAAH

• Maximal sets of nodes in which every node


can reach every other by some path

!"#"$%&'#()*+*%*,$
-./"0&'#()*+*%*,$+
-0*1*$'.&#,23'$4

5'%'&/0'6$&70,2&80,++9&:,01'%%*&;&<'0="0&>??@A
Walks, Trails, Paths

• Walk = a sequence of nodes that can be


visited by following edges
• Trail = walk with no repeated lines
• Path = walk with no repeated node
Seven Bridges of
Königsberg
Path Length & Distance

• Length of path = number of links


• Length of shortest path between two nodes =
distance or “geodesic”
• Longest geodesic between any two nodes
• = graph diameter
!"#$%&'(')*+%,#-"
Example
/0','1,%&'*+ !*
!"
/0'6*#7+
!!
"'4"%8""#'%8/ (
'6"#$%&'/0 )
%'1,%&':,7,
-; "
'

# !
&

$ %
Cutpoints
!"#$%&'#(
• Nodes, if deleted, would disconnect the
network
) *%+,(-./&0/1-&2-+,3,#,+1-.%"3+-+&(0%'',0#-',#
• Cutset = set of nodes required to keep a
graph connected
!$**"&
!$%
!"++
!"##
!&'')

!&'()
Bridges
!"#$%&'()*+,&"-&.,+(,
• An edge, if removed, would disconnect the
network 0 1&2),&23$2&#"44,#25&4"*,5&23$2&6"7
• "23,(6)5,&8,&$2&%,$52&9&52,:5&$:$(
Local bridge: connects nodes that otherwise
would be far removed
1
'
!"#$%&'&($')&*+$,-
Centralization

./,*&,'0%#1)12*,/*1$)'#&&*1)(2

• Degree to which network revolves around a


single node

<&/,'! 3/*/'4$5,*&26'$7'8149/&:';1)-
<&/,'=
Next Time

• Centrality and Power in Social Networks


• Identification of Key Players

You might also like