Introduction to the Pokemon data

Unsupervised Learning in R

Hank Roark

Senior Data Scientist at Boeing

"Real" data exercise

pokemon logo

Unsupervised Learning in R

The Pokemon dataset

head(pokemon)
HitPoints Attack Defense SpecialAttack SpecialDefense Speed
[1,] 45 49 49 65 65 45
[2,] 60 62 63 80 80 60
[3,] 80 82 83 100 100 80
[4,] 80 100 123 122 120 80
[5,] 39 52 43 60 50 65
[6,] 58 64 58 80 65 80
1 https://www.kaggle.com/abcsds/pokemon 2 https://pokemondb.net/pokedex
Unsupervised Learning in R

Data challenges

  • Selecting the variables to cluster upon
  • Scaling the data (will handle in last chapter)
  • Determining the number of clusters
    • Often no clean "elbow" in scree plot
    • This will be a core part of the exercises
  • Visualize the results for interpretation
Unsupervised Learning in R

Let's practice!

Unsupervised Learning in R

Preparing Video For Download...