Classification sorts things into groups. Regression is different — it predicts an actual
number! 📚 How many marks? 🏠 What price? The AI looks at past data, learns how changing one thing changes
the number, draws a straight trend line, and uses it to predict. Fit the line, then make a prediction!
Pick an example
✏️ Move the line to fit the dots
How close? 🎯
0%
🔮 Now predict a NEW number
🤖 In a real AI — this is regression
A regression model looks at lots of past examples and learns the best trend line all by itself (it nudges the
line until the gaps to the dots are as tiny as possible). Then it predicts a number for something brand-new:
a house's price from its size, your next test marks from study time, tomorrow's temperature, and
more. Real models use many clues at once and can bend the line — but the idea is exactly this:
the aim is to predict a numerical value.