Machine learning is a way of building software that learns patterns from examples instead of following rules written by a programmer. Show a system thousands of labelled photos and it learns to recognise a cat; show it years of sales data and it learns to forecast next month.
Traditional programming vs machine learning
In traditional programming you write the rules and give the computer data; it produces answers. In machine learning you give the computer data and answers; it works out the rules. That flip is why ML is useful for problems where the rules are too messy to write by hand, like spotting spam or understanding speech.
The three main types
- Supervised learning — learns from labelled examples. Used for spam filters, price prediction and medical image classification.
- Unsupervised learning — finds structure in unlabelled data. Used for customer segmentation and anomaly detection.
- Reinforcement learning — learns by trial and reward. Used in game-playing systems, robotics and recommendation tuning.
Six everyday examples
- Your email’s spam folder
- Netflix and YouTube recommendations
- Fraud alerts from your bank
- Voice assistants understanding what you said
- Google Maps predicting arrival times
- Photo apps grouping pictures by face
Where to start learning
You do not need advanced maths to begin. Start with a no-code tool such as Teachable Machine to train a simple image classifier, then move to a beginner Python course once you understand the workflow: collect data, train, test, improve.