What is machine learning?
Machine learning is a branch of computing where algorithms learn patterns from historical data to make predictions or decisions without manual rules.
What it means
Traditional software follows explicit rules written by a programmer: if temperature exceeds ninety degrees, trigger the alarm. Machine learning flips that setup. You feed an algorithm thousands of historical examples (inputs paired with outcomes, or raw datasets to cluster) and let mathematical optimisation fit a model that generalises to new, unseen cases.
It covers everything from simple linear regressions and decision trees up to deep neural networks and large language models.
Why it matters if you're a student
When everyone at work says "we should use AI for this", the sharpest person in the room asks whether classical machine learning solves it faster, cheaper and more transparently than a chatbot:
→ predicting next week's supermarket bread spoilage needs clean tabular machine learning, not a conversational model → smaller statistical models are far easier to audit for regulators and engineers → garbage historical data ruins every machine learning system regardless of algorithm choice
An everyday example
An energy grid operator wants to forecast wind farm generation twelve hours ahead. Instead of writing hand-crafted weather equations alone, a machine learning model trains on years of turbine output, wind speed, air pressure and seasonal patterns to predict megawatt supply hour by hour.
Common misconception
Because generative AI dominates headlines, people forget that classical machine learning still drives the vast majority of measurable business value in logistics, credit scoring, fraud detection, retail pricing and precision agriculture, at a tiny fraction of the compute cost.