1 of 25

PLEASE

INSTALL

GEPHI

http://gephi.github.io/

Mac Problems?: http://bit.ly/1OktlQP

PC Problems?: http://bit.ly/1WM1mLf

2 of 25

SOCIAL NETWORK�ANALYSIS �(methods)�

3 of 25

WHO IS THE

MOST INFLUENTIAL

PERSON ON

TWITTER?

4 of 25

76.2M Followers, Following 157

68.1M Followers, Following 236k

64.7M Followers, Following 640k

64.4M Followers, Following 244

5 of 25

A SIMPLE SOCIAL NETWORK

NODE

EDGE

6 of 25

SOCIAL

NETWORK

def. A set of relationships, often defined as a set of “nodes” or individuals, and the ties or “edges” between them

7 of 25

DIRECTED EDGE

UNDIRECTED EDGE

8 of 25

TYPES OF DYADS

No Tie/Edge

Asymmetric Edge

Symmetric Tie/Edge

9 of 25

DEGREE

def. The total number of edges

connected to each node in a network

10 of 25

IN-DEGREE

def. The total number of edges

with arrows that point towards a node in a network.

11 of 25

OUT-DEGREE

def. The total number of edges

with arrows that point away a node in a network.

12 of 25

Very High In-Degree

Very Low Out-Degree

Very High In-Degree

Very High Out-Degree

13 of 25

14 of 25

CENTRALITY

MEASURES

15 of 25

The person

with the most

followers is the

most important

The person

who follows the

most people is the

most important

The person

who is in between

The largest

clusters of people

Is the most important

The person

who is in the

Middle of the

largest cluster

is the most

important

16 of 25

Node Size=

Higher

Closeness

Centrality

Color=

Degree

Centrality

17 of 25

Which nodes are connected

to other nodes that are

themselves most influential?

18 of 25

LET’S MEASURE

THE CENTRALITY

OF CELEBRITIES

ON TWITTER

19 of 25

http://bit.ly/2lcvBC9

Download the Network Data for Gephi Here:

20 of 25

…AND THE WINNER IS:

21 of 25

WHAT CREATES

CLUSTERING

WITHIN

NETWORKS?

22 of 25

HOMOPHILY

def. The tendency for people to have network ties to those who are very similar to themselves (i.e. “birds of a feather flock together”).

23 of 25

24 of 25

25 of 25