Learning Directory:
Content directory:
Unsupervised learning includes algorithms:
Clustering: K-means (K-means clustering)
dimensionality reduction: PCA
K-means principle
API:
Case:
How to evaluate unsupervised learning?
Summary of K-means algorithm:
Features : It uses an iterative algorithm, which is intuitive, easy to understand and practical.
Disadvantages : easy to converge to a local optimal solution (when k initial points are clustered together). Multiple clustering can be used to solve the
application scenario : when there is no target value, Do clustering first, then how to classify