Social Network Analysis

Social Networks
• A structural approach to understanding social interaction.
• Networks consist of Actors and the Ties between them.

• We represent social networks

as graphs whose vertices are
the actors and whose edges
are the ties.

• Edges are usually weighted to

show the strength of the tie.
Social Networks
 A social network site allows people who share interests to build a
‘trusted’ network/ online community.

 A social network site will usually provide various ways for users to
interact, such as IM (chat/ instant messaging), email, video sharing, file
sharing, blogging, discussion groups, etc.
Social Networks
 Most contain libraries/ directories of some categories, such as former
classmates, old work colleagues, and so on (like Facebook, Linked in,
etc).

 They provide a means to connect with friends (by allowing users to

create a detailed profile page), and recommender systems linked to trust.
Social Network Concepts
 Actor
 An “actor” is a basic component for SNs. Actors can be:
 Individual people, Corporations, Nation-States, Social groups
 Modes
 If all the actors are of the same type, the network is called a one-mode
network. If there are two groups of actor then it is a two-mode network.
 E.g. an affiliation network is a two-mode network. One mode is individuals,
the other is groups to which they belong. Ties represent the relation: person A
is a member of group B.
 Ties
 A tie is the relation between two actors. Common types of ties include:
 Friendship, Amount of communication, Goods exchanged, Familial relation
(kinship), Institutional relations
Why are we studying Social Networks ?
 Businesses use SNA to analyse and improve communication flow in their
organization, or with their networks of partners and customers
 Law enforcement agencies use SNA to identify criminal and terrorist networks from
traces of communication that they collect; and then identify key players in these
networks
 Social Network Sites like Facebook use basic elements of SNA to identify and
recommend potential friends based on friends-of-friends
 Network operators (telephony, cable, mobile) use SNA-like methods to optimize the
structure and capacity of their networks.
Why are we studying Social Networks ?
 When you want to visualize your data so as to uncover patterns in relationships or
interactions
 When you want to follow the paths that information (or basically anything) follows
in social networks
 When you do quantitative research, although for qualitative research a network
perspective is also valuable
 A quantitative analysis of a social network can help you identify different types of actors in the
network or key players, whom you can focus on for your qualitative research.
Social Networks and KMS

Social
Networks

What ties Information Architecture, Knowledge Management

and Social Network Analysis more closely together is the
reciprocal relationship between people and content.
Knowledge
Information
Management
Architecture Systems
Social Network Analysis
 Social network analysis [SNA] is the mapping and measuring of relationships and flows
between people, groups, organizations, computers or other information/ knowledge processing
entities

 The nodes in the network are the people and groups while the links show relationships or flows
between the nodes
Social Network Analysis – Basic Concepts
 Networks – How to represent Networks
 Tie Strength – How to identify strong/weak ties in network
 Key Players – How to identify central nodes in network
Networks – How to represent Networks
Networks – How to represent Networks
Ties – Adding weights
Centrality in Social Networks
Intuitively, we want a method that allows us to distinguish “important” actors.

Consider the following graphs:

Centrality
In social networks, some people, like celebrities and politicians have a lot
of followers and can propagate information easier than ordinary subjects

Hence, these nodes can be considered as central

However, this definition of centrality is not unique, since we can define it

in terms of the load that each node receives.
Centrality
For example, in a street network, an urban region is central whether it
presents traffic jams and is more accessed than other places

Thus, the definition of centrality is not general and depends on the

application

Since we do not have a consensus about the general definition of

centrality, several measures have been proposed, where each one
considers specific concepts
Centrality Measures
 Degree

 Closeness

 Harmonic

 Betweenness
Degree Centrality
• Degree centrality is one of the easiest to calculate

• The degree centrality of a node is simply its degree—the number of

edges/links it has.

• The higher the degree, the more central the node is

• This can be an effective measure, since many nodes with high degrees
also have high centrality by other measures.
Degree Centrality
In Figure, node P has the highest degree
centrality of 9.

Meanwhile, node F has a relatively low

degree centrality of 5.

Many other nodes have that same

centrality value or higher (e.g., node D
has a degree centrality of 5).
Degree Centrality

Highest the score, highest the Degree centrality

Closeness centrality
• Closeness centrality indicates how close a node is to all other nodes
in the network.

• It is calculated as the average of the shortest path length from the

node to every other node in the network.
Closeness centrality
Let’s start by computing the average shortest path length of node D.
Table shows each node and the length of the shortest path from D.
Closeness centrality
Closeness is based on the inverse of the distance of the actor (in question) to every
other actor in the network.

Closeness (D)= 1/(3+2+1+1+2+2+1)= 1/12= 0.083

Closeness centrality
Closeness centrality is based on the inverse of the distance of the actor (in question)
to every other actor in the network.

Closeness (A)= 1/(1+2+3+4+5+5+4)= 1/24= 0.042

Closeness centrality Closeness Centrality in the examples
Distance Closeness normalized

0 1 1 1 1 1 1 1 .143 1.00
1 0 2 2 2 2 2 2 .077 .538
1 2 0 2 2 2 2 2 .077 .538
1 2 2 0 2 2 2 2 .077 .538
1 2 2 2 0 2 2 2 .077 .538
1 2 2 2 2 0 2 2 .077 .538
1 2 2 2 2 2 0 2 .077 .538
1 2 2 2 2 2 2 0 .077 .538

Distance Closeness normalized

0 1 2 3 4 4 3 2 1 .050 .400
1 0 1 2 3 4 4 3 2 .050 .400
2 1 0 1 2 3 4 4 3 .050 .400
3 2 1 0 1 2 3 4 4 .050 .400
4 3 2 1 0 1 2 3 4 .050 .400
4 4 3 2 1 0 1 2 3 .050 .400
3 4 4 3 2 1 0 1 2 .050 .400
2 3 4 4 3 2 1 0 1 .050 .400
1 2 3 4 4 3 2 1 0 .050 .400
Closeness centrality
Closeness Centrality in the examples
Distance Closeness normalized
0 1 2 3 4 5 6 .048 .286
1 0 1 2 3 4 5 .063 .375
2 1 0 1 2 3 4 .077 .462
3 2 1 0 1 2 3 .083 .500
4 3 2 1 0 1 2 .077 .462
5 4 3 2 1 0 1 .063 .375
6 5 4 3 2 1 0 .048 .286
Closeness Centrality

Highest the score, highest the Closeness centrality.

Harmonic Centrality
Harmonic centrality of an actor (node) u is the sum of the reciprocal of the shortest
path distances from all other actors to u.

where d(v, u) is the shortest-path distance between v and u.

Closeness centrality
Harmonic centrality of an actor (node) u is the sum of the reciprocal of the shortest
path distances from all other actors to u.

Closeness (D)=
(1/3)+(1/2)+(1/1)+(1/1)+(1/2)+(1/2)+(1/1) = 4.83
Harmonic Centrality
Harmonic centrality of an actor (node) u is the sum of the reciprocal of the shortest
path distances from all other actors to u.

Closeness (A)=
(1/1)+(1/2)+(1/3)+(1/4)+(1/5)+(1/5)+(1/4) = 2.7
Harmonic Centrality

Highest the score, highest the Harmonic centrality.

Betweenness Centrality

Model based on communication flow: A person who lies on communication paths can
control communication flow, and is thus important.

Betweenness centrality counts the number of shortest paths between i and k that
actor j resides on.
Betweenness Centrality
Betweenness centrality measures how important a node is to the shortest paths through
the network

1. To compute betweenness for a node N, we select a pair of nodes and find all the
shortest paths between those nodes
2. Then we compute the fraction of those shortest paths that include node N
3. If there were five shortest paths between a pair of nodes, and three of them went
through node N, then the fraction would be 3÷5=0.6
4. We repeat this process for every pair of nodes in the network
5. We then add up the fractions we computed, and this is the betweenness centrality
for node N
Betweenness Centrality
Let’s compute betweenness centrality for node B.

There are 10 pairs of nodes to consider: AC, AD, AE, AF, CD,
CE, CF, DE, DF, and EF.
Without counting, we know that 100% of the shortest paths from
A to every other node in the network go through B, since A can’t
reach the rest of the network without B. Thus, the fractions for
AC, AD, AE, and AF are all 1.
1. We repeat this process for every pair of nodes in the
network.
2. We then add up the fractions we computed, and this is the
betweenness centrality for node N.
Betweenness Centrality

From C to D, there are two shortest paths: one through

B and one through E. Thus, 1÷2=0.5 go through B.

For the remaining pairs—CE, CF, DE, DF, and EF—no

shortest paths go through B.
Thus, the fraction for all of these is zero.
Now we can calculate the betweenness for B:
4×1 (A to all others)+0.5 (CD)+5×0 (all remaining
pairs)=4+0.5+0=4.5
 Betweenness Centrality

In contrast, the betweenness centrality of A is zero,

since no shortest paths between D, C, D, E, and F
go through A.

Highest the score, highest Betweenness Centrality.

Which is best centrality measure?
Degree Centrality
Degree centrality is an important component of any attempt to determine
the most important people in a social network.

For example, in BrandWatch’s most influential men and women on Twitter

2017 the top 5 people in each category have over 40m followers each,
which is a lot higher than the average degree.
Degree Centrality
Degree centrality, however, can be
deceiving, because it is a purely
local measure.
Degree Centrality
Although the degree definition is intuitive, since it is
expected that highly connected nodes are at the center
of the network, it has some drawbacks.

For instance, as we can see in Figure, the nodes with the

highest degree(in black) are at the periphery of the
network and, therefore, are not central.

Thus, the degree centrality can be considered as a local

centrality measure, as a hub (densely connected node)
may not be central.
Degree Centrality

Degree centrality is a good measure of the total connections a node has,

but will not necessarily indicate the importance of a node in connecting
others or how central it is to the main group.
 Each centrality measure highlights a different aspect of a
node's importance within a network.
Betweenness
Degree Centrality Closeness Centrality Harmonic Centrality
Centrality
• This is the simplest •This measure calculates •Similar to closeness •This measure identifies
measure, indicating how how close a node is to centrality, harmonic nodes that act as
many direct all other nodes in the centrality measures a "bridges" between
connections a node has. network by considering node's proximity to different parts of the
• A node with a high the average shortest path other nodes. network by calculating
degree centrality is length between it and • It gives less weight to how often a node
considered highly every other node. very long distances, appears on the shortest
connected and • A node with high making it more robust paths between other
potentially influential closeness centrality is to outliers and pairs of nodes.
within the network. considered central and potentially better suited •A node with high
can quickly reach other for networks with betweenness centrality
nodes in the network. uneven distances has significant control
between nodes. over information flow
between different groups
within the network.
Reference
 Stanley, Wasserman, and Faust Katherine. "Social
network analysis: Methods and applications." Cambridge:
Cambridge University (1994).