10-12. February 2018
Recomputed Jaccard index on MB/CiteB and determined islands - got different results than in 2017 (other network? Cite, preprint transformation (see names), cleansing?). Checked the procedure test.
> setwd("C:/Users/batagelj/work/Python/WoS/BM/results/jaccard.new")
> C <- read.csv("islands5-75all.clu",header=FALSE,skip=1)[[1]]
> length(C)
[1] 13540
> L <- table(C)
> head(L)
C
0 1 2 3 4 5
11799 6 6 10 7 18
> F <- table(L[2:length(L)])
> F
5 6 7 8 9 10 11 12 13 14 15 17 18 24 27 28 31 33 34 40 43 48 51 52 55 59 70 71 75
33 16 11 17 12 8 4 2 2 3 1 4 2 1 1 1 2 1 1 1 1 1 1 2 1 1 1 1 1