What is supervised learning used for?#
This approach automates the process of making predictions by training an AI model on past examples. This allows businesses to build systems that can instantly categorise new data, forecast future outcomes, and identify risks with high accuracy and at a massive scale.
How does supervised learning work?#
The process begins by feeding the AI a large dataset of historical examples where the correct outcome is already known. For example, to build a house price predictor, you would provide the AI with data on thousands of houses, each including its features (square footage, number of bedrooms) and its final sale price (the "label"). The AI studies these examples to find the underlying patterns connecting the features to the final price.
Once trained, this model can be given data for a new house it has never seen before, and it will use the patterns it learned to predict a sale price. The term "supervised" refers to this process of using a labelled dataset with known answers to guide, or supervise, the AI's learning process.
How is supervised learning different from unsupervised learning?#
Supervised learning uses labelled data to train a model for a specific predictive task. In contrast, unsupervised learning is used when you don't have labelled data. Instead of predicting a known outcome, it sifts through the data to find hidden patterns or groupings on its own, such as identifying distinct segments of customers based on their purchasing behavior without being told what those segments are in advance.