Who is important in the conversation?
Different measures of centrality all try to get at the similar concept of "which vertices are most important." As we discussed earlier, these two metrics approach it slightly differently. Keep in mind that while each may give a similar distribution of centrality measures, how an individual vertex ranks according to both may be different. Now we're going to compare the top ranking vertices of Twitter users.
The vectors that store eigen and betweenness centrality are stored respectively as retweet_ec
and retweet_btw
.
This exercise is part of the course
Case Studies: Network Analysis in R
Exercise instructions
- Calculate the
0.99
quantile of the betweenness,retweet_btw
. - Subset
retweet_btw
for values greater than this quantile to keep the top 1%. - Do the same for eigen-centrality,
retweet_ec
. - Run the code that puts these in a data frame and look at the results.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Get 0.99 quantile of betweenness
betweenness_q99 <- quantile(___, ___)
# Get top 1% of vertices by betweenness
top_btw <- ___[retweet_btw > ___]
# Get 0.99 quantile of eigen-centrality
eigen_centrality_q99 <- ___(___, ___)
# Get top 1% of vertices by eigen-centrality
top_ec <- ___
# See the results as a data frame
data.frame(
Rank = seq_along(top_btw),
Betweenness = names(sort(top_btw, decreasing = TRUE)),
EigenCentrality = names(sort(top_ec, decreasing = TRUE))
)