Applying Kmeans to high dimensional data with 63 column fields.
-
Data Preparation
a. Describing the Data
b. Scaling Data: Standardization
c. Dimension Reduction
Applied PCA(Prinicpal Component Analysis) and converted all column fields into 2 independent fields.
-
Optimum K-Value
a. Silhouette Scores
b. Elbow Method
-
Fitting the finalized K-Means
-
Plotting the cluster chart