Unsupervised Learning

Unlike supervised learning, we are given a dataset with only the input features $X_j$ and no response variable $Y$.

One of the challenges of unsupervised learning is that, unlike supervised learning, it is subjective and often without a clear goal.

However, unsupervised learning thrives in that, in reality, unlabeled data is more easily obtained than labeled data.

Table of contents
  1. Standardizing Features
  2. PCA
  3. Clustering

Standardizing Features

In supervised learning, some learning methods are invariant to scaling.

However, in unsupervised learning, standardizing features is always crucial.

PCA

Principle Component Analysis (PCA)

Clustering

Clustering