Supervised Learning

• This type uses data that has been labeled.
  (About 80% of machine learning uses this.)
• It helps computers learn to make predictions.
• Some common jobs for supervised learning are:
  • Classification: like telling if an email is spam or not.
  • Regression: like figuring out prices for products.

---

Unsupervised Learning

• This type works with data that isn’t labeled.
  (It’s around 20% of applications.)
• It looks for patterns and groups in the data.
• Some tools for unsupervised learning are:
  • K-means clustering: This groups data into different clusters.
  • PCA: This helps simplify data by reducing its dimensions.

---

Impact on Data Analysis

• Supervised learning is all about being accurate in predictions.
  It can often reach over 90% accuracy in many tasks.
• Unsupervised learning helps us explore data better.
  It can show new ideas or ways to improve how we collect data.