Network centrality measures

Analyzing Social Media Data in R

Sowmya Vivek

Data Science Coach

Lesson overview

Concept of network centrality measures
Degree centrality and betweenness
Identify key players in the network and their role in a promotional campaign

Network centrality measures

Influence of a vertex is determined by the number of edges and its position
Network centrality is the measure of importance of a vertex in a network
Network centrality measures assign a numerical value to each vertex
Value is a measure of a vertex's influence on other vertices

Degree centrality

Simplest measure of vertex influence
Determines the edges or connections of a vertex
In a directed network, vertices have out-degree and in-degree scores

Out-degree

Out_degree

In-degree

In_degree

Degree centrality of a user

library(igraph)
# Calculate out-degree
out_deg <- degree(nw_rtweet, 
                  "OutfitAww", 
                   mode = c("out"))
out_deg

OutfitAww 
    20

library(igraph)
# Calculate in degree
in_deg <- degree(nw_rtweet, 
                 "OutfitAww", 
                  mode = c("in"))
in_deg

OutfitAww 
   23

Users who retweeted most

# Calculate the out-degree scores
out_degree <- degree(nw_rtweet, mode = c("out"))

# Sort the users in descending order of out-degree scores
out_degree_sort <- sort(out_degree, decreasing = TRUE)

Users who retweeted most

# View the top 3 users
out_degree_sort[1:3]

VanesEtim   RedNileShop     w3daily 
  209           147            62

Users whose posts were retweeted most

# Calculate the in-degree scores
in_degree <- degree(nw_rtweet, mode = c("in"))

# Sort the users in descending order of in-degree scores
in_degree_sort <- sort(in_degree, decreasing = TRUE)

Users whose posts were retweeted most

# View the top 3 users
in_degree_sort[1:3]

XyC_129    SocialBflyMag      jisoupy 
  171           167             142

Betweenness

Degree to which nodes stand between each other
Captures user role in allowing information to pass through network
Node with higher betweenness has more control over the network

Betweenness

Identifying users with high betweenness

# Calculate the betweenness scores of the network
betwn_nw <- betweenness(nw_rtweet, directed = TRUE)

# Sort the users in descending order of betweenness scores
betwn_nw_sort <- betwn_nw %>%
                   sort(decreasing = TRUE) %>%
                   round()

Identifying users with high betweenness

# View the top 3 users
betwn_nw_sort[1:3]

GuruOfficial   Home_and_Loving    SimplyTasheena 
   65                54                40

Let's practice!

Analyzing Social Media Data in R