What Is Machine Learning?

Machine learning is the branch of AI that most business applications are actually built on. The name describes exactly what it does: machines that learn.

Traditional software is programmed. You write the rules, the machine follows them. Machine learning flips this. You provide examples of inputs and the correct outputs, and the system learns the rules itself.

A simple example: you want to build a system that predicts which customers are likely to churn. With traditional software, you would write rules, if a customer has not logged in for 30 days and their last support ticket was unresolved, flag them. Machine learning takes a different approach. You feed it historical data, thousands of customers, their behaviour patterns, and whether they churned, and the system discovers which combinations of factors actually predicted churn. It often finds patterns a human would not think to look for.

Three types worth knowing

Supervised learning is the most common. You provide labelled examples, this email is spam, this one is not, and the system learns to classify new examples. Most business AI is supervised learning.

Unsupervised learning finds patterns in data without being told what to look for. Useful for customer segmentation, grouping customers by behaviour without pre-defining the groups.

Reinforcement learning learns by trial and error, receiving feedback on each action. Used in more complex systems like recommendation engines and robotics.

The practical takeaway

Most business machine learning is supervised learning applied to prediction problems, which leads will convert, which invoices will be paid late, which customers are at risk. If you have historical data and a question you want to answer, machine learning is often a direct path to an answer.