hierarchical cl u stering
play

Hierarchical cl u stering N E TW OR K AN ALYSIS IN TH E TIDYVE R - PowerPoint PPT Presentation

Hierarchical cl u stering 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 NETWORK ANALYSIS IN THE TIDYVERSE The similarit y meas u re


  1. Hierarchical cl u stering 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. NETWORK ANALYSIS IN THE TIDYVERSE

  4. The similarit y meas u re Single - linkage : the similarit y bet w een t w o gro u ps is the ma x im u m of the similarities bet w een nodes of di � erent gro u ps . Complete - linkage : the similarit y bet w een t w o gro u ps is the minim u m of the similarities bet w een nodes of di � erent gro u ps . A v erage - linkage : the similarit y bet w een t w o gro u ps is the a v erage of the similarities bet w een nodes of di � erent gro u ps . NETWORK ANALYSIS IN THE TIDYVERSE

  5. The cl u stering algorithm 1. E v al u ate the similarit y meas u res for all node pairs . 2. Assign each node to a gro u p of its o w n . 3. Find the pair of gro u ps w ith the highest similarit y and join them together into a single gro u p . 4. Calc u late the similarit y bet w een the ne w composite gro u p and all others . 5. Repeat steps 3 and 4 u ntil all nodes ha v e been joined into a single gro u p . NETWORK ANALYSIS IN THE TIDYVERSE

  6. Hierarchical cl u stering in R # distance matrix from similarity matrix D <- 1-S # distance object from distance matrix d <- as.dist(D) # average-linkage clustering method cc <- hclust(d, method = "average") # cut dendrogram at 4 clusters hclust(d, method = "average") [1] 1 1 1 2 2 2 1 1 1 1 1 1 1 1 1 1 1 2 2 3 2 2 2 1 4 2 2 2 [29] 2 2 2 3 2 2 4 1 1 2 2 2 1 3 1 1 2 3 1 1 4 4 1 1 1 4 1 2 [57] 3 3 3 3 3 1 1 3 NETWORK ANALYSIS IN THE TIDYVERSE

  7. Let ' s cl u ster o u r net w ork ! N E TW OR K AN ALYSIS IN TH E TIDYVE R SE

  8. Interacti v e v is u ali z ations w ith v isNet w ork 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)

  9. Different la y o u ts NETWORK ANALYSIS IN THE TIDYVERSE

  10. NETWORK ANALYSIS IN THE TIDYVERSE

  11. NETWORK ANALYSIS IN THE TIDYVERSE

  12. NETWORK ANALYSIS IN THE TIDYVERSE

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

  14. Congrat u lations ! 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)

  15. Deeper inside net w ork science Yo u no w kno w ho w to : Anal yz e an y net w ork w ith basic centralit y and similarit y meas u res Prod u ce bea u tif u l net w ork v is u ali z ations , incl u ding interacti v e ones For more information : Uni v ersit y of Udine Net w ork Science Co u rse NETWORK ANALYSIS IN THE TIDYVERSE

  16. Contin u e the jo u rne y! N E TW OR K AN ALYSIS IN TH E TIDYVE R SE

Recommend


More recommend