bet w eenness on ties
play

Bet w eenness on ties N E TW OR K AN ALYSIS IN TH E TIDYVE R SE - PowerPoint PPT Presentation

Bet w eenness on ties N E TW OR K AN ALYSIS IN TH E TIDYVE R SE Massimo Franceschet Prof . of Data Science , Uni v ersit y of Udine ( Ital y) NETWORK ANALYSIS IN THE TIDYVERSE Weighted bet w eenness NETWORK ANALYSIS IN THE TIDYVERSE Comp u


  1. Bet w eenness on ties N E TW OR K AN ALYSIS IN TH E TIDYVE R SE Massimo Franceschet Prof . of Data Science , Uni v ersit y of Udine ( Ital y)

  2. NETWORK ANALYSIS IN THE TIDYVERSE

  3. Weighted bet w eenness NETWORK ANALYSIS IN THE TIDYVERSE

  4. Comp u ting bet w eenness # compute distance weights for ties dist_weight = 1 / E(g)$weight # compute weighted betweenness on ties edge_betweenness(g, weights = dist_weight) NETWORK ANALYSIS IN THE TIDYVERSE

  5. Let ' s start practicing w ith tie bet w eenness ! N E TW OR K AN ALYSIS IN TH E TIDYVE R SE

  6. Vis u ali z ing centralit y meas u res N E TW OR K AN ALYSIS IN TH E TIDYVE R SE Massimo Franceschet Prof . of Data Science , Uni v ersit y of Udine ( Ital y)

  7. Vis u ali z ing bet w eenness # visualize the network with tie transparency proportional to betweenness ggraph(g, layout = "with_kk") + geom_edge_link(aes(alpha = betweenness)) + geom_node_point() NETWORK ANALYSIS IN THE TIDYVERSE

  8. Vis u ali z ing w eight and degree # visualize tie weight and node degree ggraph(g, layout = "with_kk") + geom_edge_link(aes(alpha = weight)) + geom_node_point(aes(size = degree)) NETWORK ANALYSIS IN THE TIDYVERSE

  9. Let ' s practice ! N E TW OR K AN ALYSIS IN TH E TIDYVE R SE

  10. The strength of w eak ties N E TW OR K AN ALYSIS IN TH E TIDYVE R SE Massimo Franceschet Prof . of Data Science , Uni v ersit y of Udine ( Ital y)

  11. NETWORK ANALYSIS IN THE TIDYVERSE

  12. Weak ties Weak ties are relationships bet w een members of di � erent comm u nities . The y lead to a di v ersit y of ideas NETWORK ANALYSIS IN THE TIDYVERSE

  13. Strong ties Strong ties are relationships bet w een people w ho li v e , w ork , or pla y together . The y lead to similar and stagnant ideas NETWORK ANALYSIS IN THE TIDYVERSE

  14. In its w eakness lies its strength Unlike con v entional armed gro u ps , w hich are o � en hierarchical and centrali z ed Large terrorist net w orks u se dispersed forms of organi z ation Balances co v ertness w ith broader operational s u pport Easier to reconstr u ct w itho u t dependencies on strong relationships NETWORK ANALYSIS IN THE TIDYVERSE

  15. Finding w eak ties # find number and percentage of weak ties ties %>% group_by(weight) %>% summarise(n = n(), p = n / nrow(ties)) %>% arrange(-n) # A tibble: 4 x 3 weight n p <int> <int> <dbl> 1 1 214 0.881 2 2 21 0.0864 3 3 6 0.0247 4 4 2 0.00823 NETWORK ANALYSIS IN THE TIDYVERSE

  16. Let ' s find w eak and strong ties in o u r net w ork ! N E TW OR K AN ALYSIS IN TH E TIDYVE R SE

Recommend


More recommend