Pengantar clustering hierarkis

Unsupervised Learning di R

Hank Roark

Senior Data Scientist at Boeing

Clustering hierarkis

  • Jumlah klaster tidak diketahui sebelumnya
  • Dua jenis: bottom-up dan top-down; kursus ini memakai bottom-up
Unsupervised Learning di R

Contoh sederhana

lima poin

Unsupervised Learning di R

Lima klaster

setiap poin adalah sebuah klaster

Unsupervised Learning di R

Empat klaster

empat klaster, satu klaster berisi dua poin

Unsupervised Learning di R

Tiga klaster

tiga klaster, dua klaster berisi dua poin dan satu berisi satu poin

Unsupervised Learning di R

Dua klaster

dua klaster, satu berisi 3 poin dan satu berisi 2 poin

Unsupervised Learning di R

Satu klaster

satu klaster mencakup semua poin

Unsupervised Learning di R

Clustering hierarkis di R

# Calculates similarity as Euclidean distance 
# between observations
dist_matrix <- dist(x)

# Returns hierarchical clustering model hclust(d = dist_matrix)
Call:
hclust(d = s)

Cluster method   : complete 
Distance         : euclidean 
Number of objects: 50
Unsupervised Learning di R

Ayo berlatih!

Unsupervised Learning di R

Preparing Video For Download...