Supervised v. Unsupervised

Supervised

Loads the model with knowledge so that we can have it predict future instances. The model is trained by data from a labeled dataset. There are two types—Classification and Regression.

  • Classification is the process of predicting a discrete class label, or category. 
  • Regression is the process of predicting a continuous value as opposed to predicting a categorical value

Unsupervised

The model is trained on an unlabed dataset and draws conclusions. These are more difficult algorithms than supervised learning since we know little to no information about the data. There are four types.

  • Dimension reduction: Reducing redundant features
  • Density estimation: Finding structure
  • Market basket analysis: If you buy a certain group of items, you're more likely to buy another group of item
  • Clustering: Grouping data points based on similarities