Text Mining and Visualization

270 Text Mining and Visualization: Case Studies Using Open-Source Tools

List of 3

$ membership: num [1:36942] 1 1 1 1 1 1 1 1 1 1 ...

$ csize : num [1:53] 36888 1 1 1 1 ...

$ no : int 53

> cl$csize

[1] 36888 1 1 1 1 2 1 1

[9] 1 1 1 1 1 1 1 1

[17] 1 1 1 1 1 1 1 1

[25] 1 1 1 1 1 1 1 1

[33] 2 1 1 1 1 1 1 1

[41] 1 1 1 1 1 1 1 1

[49] 1 1 1 1 1

From the csize attribute of our clusters variable, it is apparent that our graph consists

of one giant connected component (36,888 nodes) and 52 clusters with only one to two

nodes each. It is trivial to write a short loop, in order to check the tag names in these small

isolated clusters (only the first 10 shown here, for brevity):

# print isolated tags ("graph periphery")

l <- cl$no # number of clusters

for (i in 2:l) {

label ...