Machine Learning is a way of making computers learn patterns from data instead of programming every rule by hand.
You give a model examples, and it works out a mathematical relationship between inputs and outputs.
For example, you hand it thousands of houses along with their prices and features.
The model learns how features like area, location and number of rooms connect to price.
It can then use those learned patterns to predict prices for houses it has never seen.
Normal software is a list of rules a person wrote by hand.
Someone decides what should happen, and the computer follows the steps.
That works right up until the rules get too messy to write down.
Nobody can list every rule for what makes an email spam.
Machine learning flips it around: you show examples instead of writing rules.
You hand the computer thousands of emails already marked spam or not spam.
It finds the patterns that separate one pile from the other.
Those patterns are numbers, not sentences, so you often cannot read them.
The result is called a model, and you can point it at new emails.
The model is only ever as good as the examples you gave it.
Bad examples teach bad patterns, and the computer will not notice.
It is a guessing machine, not a thinking machine.
It has no idea what an email is, or what spam means.
So it will happily learn something silly if that is what your data shows.
Most of the work is getting the examples right, not picking the algorithm.
One question on this card. Every option explains itself once you pick, right or wrong.
55 concepts in this module, then a graded test with every answer explained.